fix(chat): 阻止页面卸载后的异步初始化修改模型选择

Reviewed-on: #15

.gitignore

@@ -14,6 +14,8 @@ backend/.env
 # 运行期生成的 SQLite 索引（vault 下的 Markdown 测试数据需提交）
 backend/data/*.db*
 backend/data/credentials/
+# 阶段验收笔记（验收用，不提交）
+backend/data/vault/验收/
 # 本机 MCP 配置、授权状态及服务器工作目录不得提交。
 backend/data/mcp/
 server.json

README.md

@@ -118,7 +118,7 @@ cd frontend
 pnpm test
 ```
 
-当前回归基线为后端 136 项测试、前端 32 项测试，且 TypeScript 类型检查和生产构建通过。测试数量会随功能增长，以本地实际输出和 CI 为准。
+当前回归基线为后端 218 项测试、前端 32 项测试，且 TypeScript 类型检查和生产构建通过。测试数量会随功能增长，以本地实际输出和 CI 为准。
 
 构建产物位于 `frontend/dist`，该目录不提交到 Git。
 

backend/app/agent/builtin_tools.py

@@ -160,10 +160,11 @@ def read_attachment(arguments: AttachmentReadArguments, _: ToolExecutionContext)
     return attachment_service.read_attachment(**arguments.model_dump())
 
 
-def transcribe_audio(arguments: AudioTranscribeArguments, _: ToolExecutionContext) -> dict:
+async def transcribe_audio(arguments: AudioTranscribeArguments, _: ToolExecutionContext) -> dict:
-    return transcription_service.create_transcription(
+    job = await transcription_service.create_transcription(
         arguments.attachment_id, arguments.language
-    ).model_dump(mode="json")
+    )
+    return job.model_dump(mode="json")
 
 
 def _register(

backend/app/benchmarks/__init__.py

@@ -0,0 +1,8 @@
+"""Benchmark 服务：RAG / Agent 数据集注册、指标计算与运行管理。
+
+模块划分：
+- metrics.py  纯函数指标（Hit@K / Recall@K / MRR / CitationHit / 分位数）
+- datasets.py 受控目录的 Dataset 注册与校验
+- rag.py      RAG Benchmark Runner（调用 retrieval.engine.search）
+- service.py  运行注册表、配置快照与报告组装
+"""

backend/app/benchmarks/datasets.py

@@ -0,0 +1,198 @@
+"""Benchmark Dataset 注册：从受控目录加载 JSON 数据集并校验。
+
+Dataset 只能来自配置目录（settings.benchmark_datasets_path），API 不接受调用方提交
+任意文件路径。目录不存在或为空时按「无数据集」处理，不报错。
+"""
+
+from __future__ import annotations
+
+import hashlib
+import json
+from dataclasses import dataclass, field
+from pathlib import Path
+
+from pydantic import BaseModel, Field, ValidationError
+
+from app.config import get_settings
+from app.contracts import (
+    BenchmarkDatasetInfo,
+    BenchmarkKind,
+    RAGDatasetCase,
+)
+from app.errors import ApiError
+
+
+@dataclass
+class RAGDataset:
+    """内存中的 RAG 数据集：元信息 + 已校验的 Case 列表 + 内容哈希。"""
+
+    dataset_id: str
+    kind: BenchmarkKind
+    version: str
+    description: str
+    cases: list[RAGDatasetCase] = field(default_factory=list)
+    content_hash: str = ""
+
+
+class _DatasetMeta(BaseModel):
+    """Dataset 元数据的最小校验模型。
+
+    list_datasets 用它逐文件校验元信息字段结构，把「合法 JSON 但字段类型错误」
+    （如 cases: 42）这类损坏文件隔离掉，而不是让 len() 抛 TypeError 拖垮整个列表。
+    """
+
+    dataset_id: str = Field(min_length=1)
+    kind: str = ""
+    version: str = ""
+    description: str = ""
+    cases: list = Field(default_factory=list)
+
+
+def _datasets_dir() -> Path:
+    return get_settings().benchmark_datasets_path
+
+
+def _dataset_files() -> list[Path]:
+    directory = _datasets_dir()
+    if not directory.is_dir():
+        return []
+    return sorted(directory.glob("*.json"))
+
+
+def _content_hash(raw: bytes) -> str:
+    return "sha256:" + hashlib.sha256(raw).hexdigest()
+
+
+def _read_json(path: Path) -> tuple[dict, bytes]:
+    """读取并解析 JSON 文件，返回 (dict, 原始字节)；非法 JSON 抛 BENCHMARK_DATASET_INVALID。"""
+    try:
+        raw_bytes = path.read_bytes()
+        return json.loads(raw_bytes.decode("utf-8")), raw_bytes
+    except (json.JSONDecodeError, OSError, UnicodeDecodeError) as exc:
+        raise ApiError(
+            422,
+            "BENCHMARK_DATASET_INVALID",
+            f"Dataset file is not valid JSON: {path.name}",
+            {"path": str(path)},
+        ) from exc
+
+
+def _dataset_from_raw(raw: dict, raw_bytes: bytes, kind: BenchmarkKind) -> RAGDataset:
+    """把单个数据集 JSON 解析为 RAGDataset，非法结构抛 BENCHMARK_DATASET_INVALID。"""
+    dataset_id = raw.get("dataset_id")
+    if not isinstance(dataset_id, str) or not dataset_id:
+        raise ApiError(
+            422,
+            "BENCHMARK_DATASET_INVALID",
+            "Dataset must declare a non-empty string 'dataset_id'.",
+            {},
+        )
+    file_kind = raw.get("kind", kind.value)
+    if file_kind != kind.value:
+        raise ApiError(
+            422,
+            "BENCHMARK_DATASET_INVALID",
+            f"Dataset kind mismatch: expected '{kind.value}', got '{file_kind}'.",
+            {"dataset_id": dataset_id},
+        )
+    raw_cases = raw.get("cases")
+    if not isinstance(raw_cases, list) or not raw_cases:
+        raise ApiError(
+            422,
+            "BENCHMARK_DATASET_INVALID",
+            "Dataset 'cases' must be a non-empty list.",
+            {"dataset_id": dataset_id},
+        )
+
+    cases: list[RAGDatasetCase] = []
+    for index, case in enumerate(raw_cases):
+        try:
+            parsed = RAGDatasetCase.model_validate(case)
+        except ValidationError as exc:
+            raise ApiError(
+                422,
+                "BENCHMARK_DATASET_INVALID",
+                f"Dataset case #{index} is invalid.",
+                {"dataset_id": dataset_id, "case_index": index, "errors": exc.errors()},
+            ) from exc
+        # 每个 Case 至少要声明一个期望 id，否则无法计算命中/召回
+        if not parsed.expected_note_ids and not parsed.expected_block_ids:
+            raise ApiError(
+                422,
+                "BENCHMARK_DATASET_INVALID",
+                f"Dataset case '{parsed.case_id}' must declare expected_note_ids or expected_block_ids.",
+                {"dataset_id": dataset_id, "case_id": parsed.case_id},
+            )
+        # citation_required=true 时必须声明 expected_block_ids，否则无法计算 Citation Hit Rate
+        if parsed.citation_required and not parsed.expected_block_ids:
+            raise ApiError(
+                422,
+                "BENCHMARK_DATASET_INVALID",
+                f"Dataset case '{parsed.case_id}' requires expected_block_ids when citation_required is true.",
+                {"dataset_id": dataset_id, "case_id": parsed.case_id},
+            )
+        cases.append(parsed)
+
+    return RAGDataset(
+        dataset_id=dataset_id,
+        kind=kind,
+        version=str(raw.get("version", "")),
+        description=str(raw.get("description", "")),
+        cases=cases,
+        content_hash=_content_hash(raw_bytes),
+    )
+
+
+def list_datasets(kind: BenchmarkKind) -> list[BenchmarkDatasetInfo]:
+    """枚举受控目录下指定 kind 的数据集元信息（不含 Case 内容）。
+
+    逐文件用 _DatasetMeta 校验元信息字段结构，单个损坏文件隔离跳过而非整体失败，
+    保证列表接口健壮；损坏细节由 load_dataset 抛出。
+    """
+    infos: list[BenchmarkDatasetInfo] = []
+    for path in _dataset_files():
+        try:
+            raw, raw_bytes = _read_json(path)
+            meta = _DatasetMeta.model_validate(raw)
+        except (ApiError, ValidationError):
+            continue
+        if meta.kind not in ("", kind.value):
+            continue
+        infos.append(
+            BenchmarkDatasetInfo(
+                dataset_id=meta.dataset_id,
+                kind=kind,
+                version=meta.version,
+                description=meta.description,
+                case_count=len(meta.cases),
+                content_hash=_content_hash(raw_bytes),
+            )
+        )
+    return infos
+
+
+def load_dataset(dataset_id: str, kind: BenchmarkKind) -> RAGDataset:
+    """按文件名加载并校验数据集；找不到抛 BENCHMARK_DATASET_NOT_FOUND。
+
+    只读取与请求 dataset_id 同名的文件（{dataset_id}.json），无关文件的损坏（JSON 语法
+    错误、UTF-8 解码错误、顶层非对象）不会阻断目标数据集加载；只有目标文件本身损坏
+    才抛 BENCHMARK_DATASET_INVALID。按现有文件 stem 精确匹配，不拼接调用方传入的路径。
+    """
+    for path in _dataset_files():
+        if path.stem != dataset_id:
+            continue
+        raw, raw_bytes = _read_json(path)
+        if not isinstance(raw, dict):
+            raise ApiError(
+                422,
+                "BENCHMARK_DATASET_INVALID",
+                "Dataset top-level must be a JSON object.",
+                {"dataset_id": dataset_id, "path": path.name},
+            )
+        return _dataset_from_raw(raw, raw_bytes, kind)
+    raise ApiError(
+        404,
+        "BENCHMARK_DATASET_NOT_FOUND",
+        f"Benchmark dataset does not exist: {dataset_id}",
+        {"dataset_id": dataset_id, "kind": kind.value},
+    )

backend/app/benchmarks/metrics.py

@@ -0,0 +1,58 @@
+"""Benchmark 指标纯函数。
+
+所有指标只依赖「按相关性降序的 retrieved id 列表」和「期望 id 集合」，不接触任何
+外部状态，便于单元测试与未来 Agent Benchmark 复用。retrieved 顺序越靠前越相关。
+"""
+
+from __future__ import annotations
+
+
+def hit_at_k(retrieved: list[str], expected: set[str], k: int) -> bool:
+    """前 k 个结果里是否命中任意期望 id（用于 Hit@1 / Hit@5）。"""
+    return any(item in expected for item in retrieved[:k])
+
+
+def recall_at_k(retrieved: list[str], expected: set[str], k: int) -> float:
+    """前 k 个结果召回的期望 id 占比；期望为空时视为 0。
+
+    结果先去重：检索结果是 Block 级，同一 Note 可能经多个 Block 重复出现，
+    直接逐项计数会把同一 Note 算多次、导致 Recall 超过 1。
+    """
+    if not expected:
+        return 0.0
+    return len(set(retrieved[:k]) & expected) / len(expected)
+
+
+def reciprocal_rank(retrieved: list[str], expected: set[str]) -> float:
+    """首个命中的倒数排名；未命中返回 0。rank 从 1 开始。"""
+    for rank, item in enumerate(retrieved, start=1):
+        if item in expected:
+            return 1.0 / rank
+    return 0.0
+
+
+def citation_hit(retrieved_block_ids: list[str], expected: set[str]) -> bool:
+    """首条结果的 block_id 是否为期望引用块（Citation Hit Rate 的逐 Case 判据）。"""
+    if not retrieved_block_ids or not expected:
+        return False
+    return retrieved_block_ids[0] in expected
+
+
+def mean(values: list[float]) -> float:
+    return sum(values) / len(values) if values else 0.0
+
+
+def percentile(values: list[float], p: float) -> float:
+    """线性插值分位数（p ∈ [0, 100]），用于 P50 / P95 延迟。空列表返回 0。"""
+    if not values:
+        return 0.0
+    ordered = sorted(values)
+    if len(ordered) == 1:
+        return ordered[0]
+    rank = (len(ordered) - 1) * (p / 100.0)
+    lo = int(rank)
+    hi = lo + 1
+    if hi >= len(ordered):
+        return ordered[-1]
+    frac = rank - lo
+    return ordered[lo] + (ordered[hi] - ordered[lo]) * frac

backend/app/benchmarks/rag.py

@@ -0,0 +1,163 @@
+"""RAG Benchmark Runner：调用检索引擎对数据集逐 Case 求值并聚合指标。
+
+只读操作，直接复用 app.retrieval.engine 的 search()，不旁路检索链路。指标按
+(mode, case, repeat) 逐样本计算，再按 mode 聚合；失败样本按零分计入质量指标分母，
+避免把执行失败误判为检索质量（同时保留 total/successful/failed/failure_rate）。
+"""
+
+from __future__ import annotations
+
+import asyncio
+import logging
+import time
+from collections.abc import Callable
+
+from app import repository
+from app.benchmarks import metrics as m
+from app.benchmarks.datasets import RAGDataset
+from app.contracts import (
+    RAGCaseResult,
+    RAGDatasetCase,
+    RAGMetrics,
+    RAGRunRequest,
+    SearchMode,
+    SearchRequest,
+)
+from app.retrieval.engine import engine
+from app.retrieval.provenance import capture_embedding
+
+logger = logging.getLogger(__name__)
+
+
+class BenchmarkCancelled(Exception):
+    """运行在 Case 之间被取消时抛出，用于中断后台执行并标记 cancelled。"""
+
+
+async def run_rag(
+    dataset: RAGDataset,
+    request: RAGRunRequest,
+    on_case: Callable[[RAGCaseResult, int, int], None] | None = None,
+    should_cancel: Callable[[], bool] | None = None,
+) -> tuple[dict[str, RAGMetrics], list[RAGCaseResult]]:
+    """执行 RAG Benchmark，返回 (按 mode 聚合的指标, 全部逐样本结果)。
+
+    on_case 在每个样本求值完成后回调 (result, done, total)，供上层更新进度与事件。
+    should_cancel 在每个样本开始前被检查；返回 True 时抛出 BenchmarkCancelled 中断运行。
+    """
+    total = len(request.modes) * len(dataset.cases) * request.repeat
+    done = 0
+    results: list[RAGCaseResult] = []
+
+    for mode in request.modes:
+        for case in dataset.cases:
+            expected_notes = _expected_notes(case)
+            for repeat in range(request.repeat):
+                # 让出事件循环：使运行中取消、SSE 进度与并发 API 请求能及时得到调度
+                await asyncio.sleep(0)
+                if should_cancel is not None and should_cancel():
+                    raise BenchmarkCancelled()
+                result = await _evaluate_one(case, mode, request, repeat, expected_notes)
+                results.append(result)
+                done += 1
+                if on_case is not None:
+                    on_case(result, done, total)
+
+    metrics_by_mode = {mode.value: _aggregate(results, mode) for mode in request.modes}
+    return metrics_by_mode, results
+
+
+def _expected_notes(case: RAGDatasetCase) -> set[str]:
+    """返回笔记级期望 id；仅标注块 ID 时从块反查所属笔记，避免把标注缺失误判为检索失败。"""
+    if case.expected_note_ids:
+        return set(case.expected_note_ids)
+    return {hit.note_id for hit in repository.get_block_hits(case.expected_block_ids)}
+
+
+async def _evaluate_one(
+    case: RAGDatasetCase,
+    mode: SearchMode,
+    request: RAGRunRequest,
+    repeat: int,
+    expected_notes: set[str],
+) -> RAGCaseResult:
+    search_request = SearchRequest(
+        query=case.query,
+        mode=mode,
+        limit=request.retrieval.top_k,
+        include_snippet=False,
+        rrf_k=request.retrieval.rrf_k,
+        rerank=request.retrieval.rerank,
+        rerank_candidates=request.retrieval.rerank_candidates,
+        score_threshold=request.retrieval.score_threshold,
+    )
+    start = time.perf_counter()
+    embedding = {}
+    try:
+        with capture_embedding() as embedding:
+            response = await engine.search(search_request)
+        latency_ms = (time.perf_counter() - start) * 1000.0
+    except Exception as exc:  # 单个样本失败不中断整个 Benchmark
+        # 详细异常只进日志，公开响应只带项目错误码与安全消息，避免泄露路径/SQL 等敏感信息
+        logger.warning(
+            "RAG case evaluation failed: case=%s mode=%s", case.case_id, mode.value,
+            exc_info=exc,
+        )
+        return RAGCaseResult(
+            embedding=embedding,
+            case_id=case.case_id,
+            mode=mode,
+            repeat=repeat,
+            latency_ms=(time.perf_counter() - start) * 1000.0,
+            citation_applicable=case.citation_required,
+            error="RAG case evaluation failed.",
+            error_code="BENCHMARK_CASE_EVALUATION_FAILED",
+        )
+
+    retrieved_note_ids = [item.note_id for item in response.items]
+    retrieved_block_ids = [item.block_id for item in response.items]
+    expected_blocks = set(case.expected_block_ids)
+    k = request.retrieval.top_k
+
+    return RAGCaseResult(
+        embedding=embedding,
+        case_id=case.case_id,
+        mode=mode,
+        repeat=repeat,
+        latency_ms=latency_ms,
+        retrieved_note_ids=retrieved_note_ids,
+        retrieved_block_ids=retrieved_block_ids,
+        hit_at_1=m.hit_at_k(retrieved_note_ids, expected_notes, 1),
+        hit_at_5=m.hit_at_k(retrieved_note_ids, expected_notes, 5),
+        recall=m.recall_at_k(retrieved_note_ids, expected_notes, k),
+        reciprocal_rank=m.reciprocal_rank(retrieved_note_ids, expected_notes),
+        citation_hit=m.citation_hit(retrieved_block_ids, expected_blocks),
+        citation_applicable=case.citation_required,
+    )
+
+
+def _aggregate(cases: list[RAGCaseResult], mode: SearchMode) -> RAGMetrics:
+    samples = [c for c in cases if c.mode == mode]
+    total = len(samples)
+    failed = sum(1 for c in samples if c.error is not None)
+    successful = total - failed
+    if total == 0:
+        return RAGMetrics()
+
+    # 延迟只统计成功样本；失败样本按零分计入质量指标分母，避免汇总虚高
+    latencies = [c.latency_ms for c in samples if c.error is None]
+    citation_samples = [c for c in samples if c.citation_applicable]
+    return RAGMetrics(
+        hit_at_1=m.mean([1.0 if (c.error is None and c.hit_at_1) else 0.0 for c in samples]),
+        hit_at_5=m.mean([1.0 if (c.error is None and c.hit_at_5) else 0.0 for c in samples]),
+        recall_at_k=m.mean([c.recall if c.error is None else 0.0 for c in samples]),
+        mrr=m.mean([c.reciprocal_rank if c.error is None else 0.0 for c in samples]),
+        citation_hit_rate=m.mean(
+            [1.0 if (c.error is None and c.citation_hit) else 0.0 for c in citation_samples]
+        ),
+        p50_latency_ms=m.percentile(latencies, 50.0),
+        p95_latency_ms=m.percentile(latencies, 95.0),
+        total_cases=total,
+        successful_cases=successful,
+        failed_cases=failed,
+        failure_rate=failed / total,
+    )

backend/app/benchmarks/service.py

@@ -0,0 +1,349 @@
+"""Benchmark 服务：运行注册表、配置快照与报告组装。
+
+RAG Benchmark 采用「创建即返回 queued、后台 Task 异步执行」的模式（与 index_service
+的 rebuild 一致）：POST 创建后立即返回 202 queued 的 BenchmarkRun，由受管 asyncio.Task
+在后台逐 Case 求值，进度与事件实时写入内存注册表，供 SSE 订阅。运行记录、事件与报告
+暂存内存（_runs/_events/_reports），不持久化到 SQLite；后续接入异步任务队列时再落库。
+"""
+
+from __future__ import annotations
+
+import asyncio
+import logging
+import sys
+from datetime import datetime, timezone
+from uuid import uuid4
+
+from app import repository
+from app.benchmarks import datasets
+from app.benchmarks.datasets import RAGDataset
+from app.benchmarks.rag import BenchmarkCancelled, run_rag
+from app.config import get_settings
+from app.contracts import (
+    BenchmarkEvent,
+    BenchmarkEventType,
+    BenchmarkKind,
+    BenchmarkReport,
+    BenchmarkRun,
+    BenchmarkStatus,
+    RAGCaseResult,
+    RAGMetrics,
+    RAGRunRequest,
+    SearchMode,
+)
+from app.errors import ApiError
+from app.retrieval.engine import engine
+
+logger = logging.getLogger(__name__)
+
+_runs: dict[str, BenchmarkRun] = {}
+_events: dict[str, list[BenchmarkEvent]] = {}
+_reports: dict[str, BenchmarkReport] = {}
+_tasks: dict[str, asyncio.Task] = {}
+_subscribers: dict[str, list[asyncio.Queue[BenchmarkEvent]]] = {}
+_cancel_flags: dict[str, asyncio.Event] = {}
+MAX_RUNS = 100
+
+
+def _now() -> datetime:
+    return datetime.now(timezone.utc)
+
+
+def _forget(run_id: str) -> None:
+    """移除一条 run 的全部内存态；仅在 run 处于终态时调用，避免打断活动任务。"""
+    _runs.pop(run_id, None)
+    _events.pop(run_id, None)
+    _reports.pop(run_id, None)
+    _tasks.pop(run_id, None)
+    _subscribers.pop(run_id, None)
+    _cancel_flags.pop(run_id, None)
+
+
+def _evict_terminal() -> bool:
+    """超过容量时淘汰最旧的终态 run；全部为活动 run 无法淘汰时返回 False。
+
+    绝不能删除仍在运行（queued/running）的 run：那会连带移除其 _cancel_flags 与
+    _subscribers，使后台 Task 访问时抛出 KeyError。
+    """
+    terminal = (BenchmarkStatus.completed, BenchmarkStatus.failed, BenchmarkStatus.cancelled)
+    while len(_runs) >= MAX_RUNS:
+        victim = next(
+            (rid for rid, run in _runs.items() if run.status in terminal), None
+        )
+        if victim is None:
+            return False
+        _forget(victim)
+    return True
+
+
+def _config_snapshot(request: RAGRunRequest, dataset: RAGDataset) -> dict:
+    """记录运行时的模型 / 索引 / 环境信息，保证报告可解释、可复现。"""
+    settings = get_settings()
+    return {
+        "dataset_id": dataset.dataset_id,
+        "dataset_hash": dataset.content_hash,
+        "dataset_version": dataset.version,
+        "modes": [m.value for m in request.modes],
+        "retrieval": request.retrieval.model_dump(),
+        "repeat": request.repeat,
+        "embedding": {"policy": "per_case", "details": "cases[].embedding"},
+        "local_embedding": {
+            "model_id": engine.embedding.model_id,
+            "version": engine.embedding.version,
+            "dim": engine.embedding.dim,
+        },
+        "reranker": {
+            "model_id": engine.reranker.model_id,
+            "version": engine.reranker.version,
+        },
+        "index_meta": repository.get_index_meta(),
+        "app": {"version": settings.version, "environment": settings.environment},
+        "python": sys.version.split()[0],
+        "metadata": request.metadata,
+    }
+
+
+async def _validate_index_compatibility(request: RAGRunRequest) -> None:
+    """创建 RAG Run 前校验索引已建立且与当前 Embedding 模型/维度兼容。
+
+    空索引或不兼容索引会让所有模式得到全 0 指标，把环境/索引错误误判为检索质量差，
+    故在创建时即拒绝，返回 BENCHMARK_INDEX_INCOMPATIBLE。
+    """
+    stats = repository.stats()
+    meta = repository.get_index_meta()
+    needs_vector = any(m in (SearchMode.vector, SearchMode.hybrid) for m in request.modes)
+
+    reasons: list[str] = []
+    if stats["blocks"] == 0:
+        reasons.append("index is empty (no indexed blocks; run /api/index/rebuild first)")
+    if needs_vector:
+        if meta.get("embedding_model") != engine.embedding.model_id:
+            reasons.append(
+                f"embedding model mismatch: index={meta.get('embedding_model')!r}, "
+                f"engine={engine.embedding.model_id!r}"
+            )
+        if meta.get("embedding_dim") != str(engine.embedding.dim):
+            reasons.append(
+                f"embedding dimension mismatch: index={meta.get('embedding_dim')!r}, "
+                f"engine={engine.embedding.dim}"
+            )
+        if await engine.vector_store.count() == 0:
+            reasons.append("vector index is empty")
+    if reasons:
+        raise ApiError(
+            409,
+            "BENCHMARK_INDEX_INCOMPATIBLE",
+            "Benchmark index is not built or is incompatible with the current retrieval engine.",
+            {"reasons": reasons},
+        )
+
+
+async def create_rag_run(request: RAGRunRequest) -> BenchmarkRun:
+    """创建一次 RAG Benchmark，立即返回 queued 的 BenchmarkRun，由后台 Task 执行。"""
+    dataset = datasets.load_dataset(request.dataset_id, BenchmarkKind.rag)
+    await _validate_index_compatibility(request)
+
+    # 容量检查：先淘汰终态 run 腾空间；满容量且全为活动 run 时拒绝创建
+    if not _evict_terminal():
+        raise ApiError(
+            429,
+            "BENCHMARK_CAPACITY_EXCEEDED",
+            "Benchmark run capacity exceeded; wait for active runs to finish.",
+            {},
+        )
+
+    run_id = "benchmark_" + uuid4().hex[:12]
+    snapshot = _config_snapshot(request, dataset)
+    run = BenchmarkRun(
+        run_id=run_id,
+        kind=BenchmarkKind.rag,
+        dataset_id=dataset.dataset_id,
+        dataset_hash=dataset.content_hash,
+        status=BenchmarkStatus.queued,
+        progress=0.0,
+        config_snapshot=snapshot,
+        created_at=_now(),
+    )
+    _runs[run_id] = run
+    _events[run_id] = []
+    _subscribers[run_id] = []
+    _cancel_flags[run_id] = asyncio.Event()
+    _tasks[run_id] = asyncio.create_task(_execute_rag(run_id, request, dataset, snapshot))
+    return run
+
+
+async def _execute_rag(
+    run_id: str, request: RAGRunRequest, dataset: RAGDataset, snapshot: dict
+) -> None:
+    """后台执行 RAG Benchmark，实时更新进度/事件，结束后写入报告并关闭订阅。"""
+    cancel_event = _cancel_flags[run_id]
+
+    def emit(event_type: BenchmarkEventType, data: dict) -> None:
+        sequence = len(_events[run_id])
+        event = BenchmarkEvent(
+            event=event_type, run_id=run_id, sequence=sequence, data=data, timestamp=_now()
+        )
+        _events[run_id].append(event)
+        for queue in _subscribers.get(run_id, []):
+            queue.put_nowait(event)
+
+    def finish() -> None:
+        _subscribers.pop(run_id, None)
+        _cancel_flags.pop(run_id, None)
+
+    _runs[run_id] = _runs[run_id].model_copy(
+        update={"status": BenchmarkStatus.running, "started_at": _now()}
+    )
+    emit(
+        BenchmarkEventType.run_started,
+        {"dataset_id": dataset.dataset_id, "modes": [m.value for m in request.modes]},
+    )
+    total = len(request.modes) * len(dataset.cases) * request.repeat
+
+    def on_case(result: RAGCaseResult, done: int, _total: int) -> None:
+        progress = done / total if total else 1.0
+        _runs[run_id] = _runs[run_id].model_copy(update={"progress": progress})
+        emit(BenchmarkEventType.case_completed, result.model_dump(mode="json"))
+
+    try:
+        metrics_by_mode, results = await run_rag(
+            dataset,
+            request,
+            on_case=on_case,
+            should_cancel=cancel_event.is_set,
+        )
+    except BenchmarkCancelled:
+        _runs[run_id] = _runs[run_id].model_copy(
+            update={
+                "status": BenchmarkStatus.cancelled,
+                "progress": 1.0,
+                "completed_at": _now(),
+            }
+        )
+        emit(BenchmarkEventType.run_cancelled, {"status": BenchmarkStatus.cancelled.value})
+        _reports[run_id] = BenchmarkReport(
+            run_id=run_id,
+            kind=BenchmarkKind.rag,
+            dataset_id=dataset.dataset_id,
+            dataset_hash=dataset.content_hash,
+            status=BenchmarkStatus.cancelled,
+            config_snapshot=snapshot,
+        )
+        finish()
+        return
+    except Exception as exc:  # 单次运行失败不拖垮服务，记录错误后结束
+        # 详细异常只进日志，公开响应仅带项目错误码与安全消息，避免泄露路径/SQL 等敏感信息
+        logger.exception("Benchmark run failed: run_id=%s", run_id)
+        _runs[run_id] = _runs[run_id].model_copy(
+            update={
+                "status": BenchmarkStatus.failed,
+                "progress": 1.0,
+                "error": "Benchmark run failed.",
+                "error_code": "BENCHMARK_RUN_FAILED",
+                "completed_at": _now(),
+            }
+        )
+        emit(
+            BenchmarkEventType.run_failed,
+            {"error": "Benchmark run failed.", "error_code": "BENCHMARK_RUN_FAILED"},
+        )
+        _reports[run_id] = BenchmarkReport(
+            run_id=run_id,
+            kind=BenchmarkKind.rag,
+            dataset_id=dataset.dataset_id,
+            dataset_hash=dataset.content_hash,
+            status=BenchmarkStatus.failed,
+            config_snapshot=snapshot,
+            error="Benchmark run failed.",
+            error_code="BENCHMARK_RUN_FAILED",
+        )
+        finish()
+        return
+
+    metrics = {mode: m.model_dump() for mode, m in metrics_by_mode.items()}
+    _runs[run_id] = _runs[run_id].model_copy(
+        update={
+            "status": BenchmarkStatus.completed,
+            "progress": 1.0,
+            "metrics": metrics,
+            "completed_at": _now(),
+        }
+    )
+    emit(BenchmarkEventType.run_completed, {"metrics": metrics})
+    _reports[run_id] = BenchmarkReport(
+        run_id=run_id,
+        kind=BenchmarkKind.rag,
+        dataset_id=dataset.dataset_id,
+        dataset_hash=dataset.content_hash,
+        status=BenchmarkStatus.completed,
+        config_snapshot=snapshot,
+        metrics=metrics,
+        cases=results,
+    )
+    finish()
+
+
+def list_runs(
+    kind: BenchmarkKind | None = None,
+    status: BenchmarkStatus | None = None,
+    limit: int = 50,
+    offset: int = 0,
+) -> tuple[list[BenchmarkRun], int]:
+    runs = list(_runs.values())
+    if kind is not None:
+        runs = [r for r in runs if r.kind == kind]
+    if status is not None:
+        runs = [r for r in runs if r.status == status]
+    runs.sort(key=lambda r: r.created_at, reverse=True)
+    total = len(runs)
+    return runs[offset : offset + limit], total
+
+
+def get_run(run_id: str) -> BenchmarkRun | None:
+    return _runs.get(run_id)
+
+
+def get_report(run_id: str) -> BenchmarkReport | None:
+    return _reports.get(run_id)
+
+
+def get_events(run_id: str) -> list[BenchmarkEvent]:
+    return _events.get(run_id, [])
+
+
+def cancel_run(run_id: str) -> BenchmarkRun | None:
+    """取消运行：对 queued/running 设置取消标志，后台 Task 在 Case 边界检查后置为 cancelled。"""
+    run = _runs.get(run_id)
+    if run is None:
+        return None
+    if run.status in (BenchmarkStatus.queued, BenchmarkStatus.running):
+        _cancel_flags[run_id].set()
+    return run
+
+
+def subscribe(run_id: str) -> asyncio.Queue[BenchmarkEvent] | None:
+    """订阅运行事件流；运行已结束（completed/failed/cancelled）时返回 None。"""
+    run = _runs.get(run_id)
+    if run is None or run.status in (
+        BenchmarkStatus.completed,
+        BenchmarkStatus.failed,
+        BenchmarkStatus.cancelled,
+    ):
+        return None
+    queue: asyncio.Queue[BenchmarkEvent] = asyncio.Queue()
+    _subscribers.setdefault(run_id, []).append(queue)
+    return queue
+
+
+def unsubscribe(run_id: str, queue: asyncio.Queue[BenchmarkEvent]) -> None:
+    subscribers = _subscribers.get(run_id)
+    if subscribers and queue in subscribers:
+        subscribers.remove(queue)
+
+
+async def wait_for_run(run_id: str) -> BenchmarkRun:
+    """等待后台任务结束（测试/轮询用）；无任务时直接返回当前状态。"""
+    task = _tasks.get(run_id)
+    if task is not None:
+        await task
+    return _runs.get(run_id)

backend/app/config.py

@@ -24,6 +24,7 @@ class Settings:
     db_path: Path
     vault_path: Path
     attachments_path: Path
+    benchmark_datasets_path: Path
 
 
 @lru_cache
@@ -41,4 +42,7 @@ def get_settings() -> Settings:
         attachments_path=Path(
             os.getenv("APP_ATTACHMENTS_PATH", str(data_dir / "attachments"))
         ),
+        benchmark_datasets_path=Path(
+            os.getenv("APP_BENCHMARK_DATASETS_PATH", str(data_dir / "benchmarks"))
+        ),
     )

backend/app/container.py

@@ -7,6 +7,7 @@ from app.config import BACKEND_DIR, get_settings
 from app.extensions import PluginRuntime, SkillRuntime
 from app.extensions.mcp_registry import McpServerRegistry
 from app.providers import MockProvider, ProviderFactory, ProviderRegistry
+from app.providers.routing import ModelRoutingService
 from app.providers.credentials import (
     ChainedCredentialResolver,
     EncryptedCredentialStore,
@@ -18,6 +19,7 @@ from app.providers.credentials import (
 class ApplicationContainer:
     providers: ProviderRegistry
     provider_factory: ProviderFactory
+    model_routing: ModelRoutingService
     credentials: EncryptedCredentialStore
     tools: ToolRegistry
     permissions: PermissionManager
@@ -33,7 +35,7 @@ def build_container() -> ApplicationContainer:
     provider_factory = ProviderFactory(
         ChainedCredentialResolver(credentials, EnvironmentCredentialResolver())
     )
-    providers = ProviderRegistry()
+    providers = ProviderRegistry(provider_factory)
     providers.register(
         ProviderConfig(
             provider_id="mock",
@@ -86,6 +88,7 @@ def build_container() -> ApplicationContainer:
     return ApplicationContainer(
         providers=providers,
         provider_factory=provider_factory,
+        model_routing=ModelRoutingService(providers, provider_factory.credentials),
         credentials=credentials,
         tools=tools,
         permissions=permissions,

backend/app/contracts.py

@@ -2,7 +2,7 @@ from datetime import datetime
 from enum import Enum
 from typing import Annotated, Any, Literal
 
-from pydantic import BaseModel, ConfigDict, Field, SecretStr
+from pydantic import BaseModel, ConfigDict, Field, SecretStr, field_validator
 
 
 class Contract(BaseModel):
@@ -144,6 +144,12 @@ class SearchRequest(Contract):
     limit: int = Field(default=20, ge=1, le=100)
     offset: int = Field(default=0, ge=0)
     include_snippet: bool = True
+    # 检索调优参数（Benchmark 与 Skill 共用）：控制 RRF / 精排 / 候选池 / 分数阈值。
+    # rerank_candidates=None 表示对全部候选精排（保留原有行为），Benchmark 传显式值。
+    rrf_k: int = Field(default=60, ge=1)
+    rerank: bool = True
+    rerank_candidates: int | None = Field(default=None, ge=1)
+    score_threshold: float = Field(default=0.0, ge=0.0)
 
 
 class Citation(Contract):
@@ -230,6 +236,8 @@ class ModelCapability(str, Enum):
     streaming = "streaming"
     structured_output = "structured_output"
     embedding = "embedding"
+    transcription = "transcription"
+    speaker_matching = "speaker_matching"
 
 
 class ModelRequest(Contract):
@@ -757,7 +765,24 @@ class ProviderType(str, Enum):
     ollama = "ollama"
 
 
-class ProviderConfig(Contract):
+class ProviderConnectionFields(Contract):
+    base_url: str | None = None
+    credential_id: str | None = None
+
+    @field_validator("base_url")
+    @classmethod
+    def provider_url(cls, value: str | None) -> str | None:
+        if value is None:
+            return value
+        from urllib.parse import urlsplit
+        parsed = urlsplit(value)
+        if (parsed.scheme not in {"http", "https"} or not parsed.hostname or
+                parsed.username or parsed.password or parsed.query or parsed.fragment):
+            raise ValueError("Base URL requires HTTP(S), without credentials, query or fragment")
+        return value.rstrip("/")
+
+
+class ProviderConfig(ProviderConnectionFields):
     provider_id: str
     provider_type: ProviderType
     name: str
@@ -768,7 +793,7 @@ class ProviderConfig(Contract):
     capabilities: list[ModelCapability] = Field(default_factory=list)
 
 
-class ProviderCreateRequest(Contract):
+class ProviderCreateRequest(ProviderConnectionFields):
     provider_type: ProviderType
     name: str
     base_url: str | None = None
@@ -777,7 +802,8 @@ class ProviderCreateRequest(Contract):
     enabled: bool = True
 
 
-class ProviderUpdateRequest(Contract):
+class ProviderUpdateRequest(ProviderConnectionFields):
+    provider_type: ProviderType | None = None
     name: str | None = None
     base_url: str | None = None
     default_model: str | None = None
@@ -796,6 +822,80 @@ class ProviderPreset(Contract):
     base_url: str
     default_credential_id: str | None = None
     requires_credential: bool = True
+    logo_id: str = "custom"
+    description: str = ""
+    capabilities: list[ModelCapability] = Field(default_factory=list)
+
+
+class ModelBinding(Contract):
+    provider_id: str = Field(min_length=1, max_length=128)
+    model: str = Field(min_length=1, max_length=256)
+    endpoint: str = Field(min_length=1, max_length=256)
+    dimensions: int | None = Field(default=None, ge=1, le=16384)
+
+    @field_validator("endpoint")
+    @classmethod
+    def relative_endpoint(cls, value: str) -> str:
+        # An endpoint is a path on the selected provider, never a second origin.
+        import re
+        if not re.fullmatch(r"/[A-Za-z0-9_/-]+", value) or value.startswith("//"):
+            raise ValueError("endpoint must be an absolute API path on the provider")
+        return value
+
+    @field_validator("model", "provider_id")
+    @classmethod
+    def non_blank(cls, value: str) -> str:
+        if not value.strip():
+            raise ValueError("value must not be blank")
+        return value.strip()
+
+
+class ModelRoutingConfig(Contract):
+    version: int = Field(default=0, ge=0)
+    embedding: ModelBinding | None = None
+    transcription: ModelBinding | None = None
+    speaker_matching: ModelBinding | None = None
+
+
+class LocalBackendStatus(Contract):
+    capability: Literal["embedding", "transcription", "speaker_matching"]
+    status: Literal["placeholder", "not_installed", "ready"]
+    message: str
+
+
+class ModelRoutingResponse(Contract):
+    config: ModelRoutingConfig
+    local_backends: list[LocalBackendStatus]
+
+
+class EmbeddingRequest(Contract):
+    texts: list[str] = Field(min_length=1, max_length=256)
+
+    @field_validator("texts")
+    @classmethod
+    def bound_texts(cls, value: list[str]) -> list[str]:
+        if sum(len(text) for text in value) > 200_000:
+            raise ValueError("embedding input is too large")
+        return value
+
+
+class EmbeddingResult(Contract):
+    vectors: list[list[float]]
+    source: Literal["api", "local"]
+    model_id: str
+    dimensions: int
+    fallback_reason: str | None = None
+
+
+class SpeakerMatchRequest(Contract):
+    attachment_id: str
+    reference_attachment_id: str
+
+
+class SpeakerMatchResult(Contract):
+    score: float = Field(ge=0, le=1, allow_inf_nan=False)
+    source: Literal["api", "local"]
+    fallback_reason: str | None = None
 
 
 class ProviderPresetListResponse(Contract):
@@ -888,9 +988,13 @@ class TranscriptionJob(Contract):
     error_code: str | None = None
     error_message: str | None = None
     created_at: datetime
+    source: Literal["api", "local", "sidecar"] | None = None
+    fallback_reason: str | None = None
 
 
 class IndexStatus(Contract):
+    total_notes: int = 0
+    total_blocks: int = 0
     status: Literal["idle", "queued", "running", "failed"] = "idle"
     pending_jobs: int = 0
     active_job_id: str | None = None
@@ -909,3 +1013,152 @@ class IndexJob(Contract):
     status: Literal["queued", "running", "completed", "failed"]
     scope: Literal["all", "notes", "vectors"]
     created_at: datetime
+
+
+# Benchmark
+class BenchmarkKind(str, Enum):
+    rag = "rag"
+    agent = "agent"
+
+
+class BenchmarkStatus(str, Enum):
+    queued = "queued"
+    running = "running"
+    completed = "completed"
+    failed = "failed"
+    cancelled = "cancelled"
+
+
+class RAGDatasetCase(Contract):
+    case_id: str
+    query: str = Field(min_length=1)
+    expected_note_ids: list[str] = Field(default_factory=list)
+    expected_block_ids: list[str] = Field(default_factory=list)
+    citation_required: bool = False
+    tags: list[str] = Field(default_factory=list)
+
+
+class RAGRetrievalConfig(Contract):
+    """RAG Benchmark 的检索参数。top_k 映射到 SearchRequest.limit，
+    其余参数透传到 SearchRequest，由检索引擎实际执行。"""
+
+    top_k: int = Field(default=10, ge=1, le=100)
+    rrf_k: int = Field(default=60, ge=1)
+    rerank: bool = True
+    rerank_candidates: int = Field(default=20, ge=1)
+    score_threshold: float = Field(default=0.0, ge=0.0)
+
+
+class RAGRunRequest(Contract):
+    dataset_id: str = Field(min_length=1)
+    modes: list[SearchMode] = Field(
+        default_factory=lambda: [SearchMode.fts, SearchMode.vector, SearchMode.hybrid],
+        min_length=1,
+    )
+    retrieval: RAGRetrievalConfig = Field(default_factory=RAGRetrievalConfig)
+    repeat: int = Field(default=1, ge=1, le=10)
+    metadata: dict[str, Any] = Field(default_factory=dict)
+
+    @field_validator("modes")
+    @classmethod
+    def _no_duplicate_modes(cls, value: list[SearchMode]) -> list[SearchMode]:
+        if len(value) != len(set(value)):
+            raise ValueError("modes must not contain duplicates")
+        return value
+
+
+class RAGMetrics(Contract):
+    hit_at_1: float = 0.0
+    hit_at_5: float = 0.0
+    recall_at_k: float = 0.0
+    mrr: float = 0.0
+    citation_hit_rate: float = 0.0
+    p50_latency_ms: float = 0.0
+    p95_latency_ms: float = 0.0
+    # 样本构成：失败样本按零分计入质量指标，汇总不虚高；报告据此可知实际分母
+    total_cases: int = 0
+    successful_cases: int = 0
+    failed_cases: int = 0
+    failure_rate: float = 0.0
+
+
+class BenchmarkDatasetInfo(Contract):
+    dataset_id: str
+    kind: BenchmarkKind
+    version: str
+    description: str = ""
+    case_count: int
+    content_hash: str
+
+
+class BenchmarkDatasetListResponse(Contract):
+    items: list[BenchmarkDatasetInfo] = Field(default_factory=list)
+
+
+class BenchmarkRun(Contract):
+    run_id: str
+    kind: BenchmarkKind
+    dataset_id: str
+    dataset_hash: str
+    status: BenchmarkStatus
+    progress: float | None = None
+    metrics: dict[str, Any] | None = None
+    config_snapshot: dict[str, Any] = Field(default_factory=dict)
+    error: str | None = None
+    error_code: str | None = None
+    created_at: datetime
+    started_at: datetime | None = None
+    completed_at: datetime | None = None
+
+
+class BenchmarkRunListResponse(Contract):
+    items: list[BenchmarkRun] = Field(default_factory=list)
+    page: PageMeta = Field(default_factory=PageMeta)
+
+
+class BenchmarkEventType(str, Enum):
+    run_started = "RunStarted"
+    case_completed = "CaseCompleted"
+    run_completed = "RunCompleted"
+    run_failed = "RunFailed"
+    run_cancelled = "RunCancelled"
+
+
+class BenchmarkEvent(Contract):
+    event: BenchmarkEventType
+    run_id: str
+    sequence: int
+    data: dict[str, Any] = Field(default_factory=dict)
+    timestamp: datetime
+
+
+class RAGCaseResult(Contract):
+    embedding: dict[str, Any] = Field(default_factory=dict)
+    case_id: str
+    mode: SearchMode
+    repeat: int
+    latency_ms: float
+    retrieved_note_ids: list[str] = Field(default_factory=list)
+    retrieved_block_ids: list[str] = Field(default_factory=list)
+    hit_at_1: bool = False
+    hit_at_5: bool = False
+    recall: float = 0.0
+    reciprocal_rank: float = 0.0
+    citation_hit: bool = False
+    # 该 Case 是否声明了 expected_block_ids（决定是否计入 citation_hit_rate 分母）
+    citation_applicable: bool = False
+    error: str | None = None
+    error_code: str | None = None
+
+
+class BenchmarkReport(Contract):
+    run_id: str
+    kind: BenchmarkKind
+    dataset_id: str
+    dataset_hash: str
+    status: BenchmarkStatus
+    config_snapshot: dict[str, Any] = Field(default_factory=dict)
+    metrics: dict[str, Any] = Field(default_factory=dict)
+    cases: list[RAGCaseResult] = Field(default_factory=list)
+    error: str | None = None
+    error_code: str | None = None

backend/app/errors.py

@@ -36,7 +36,11 @@ async def validation_error_handler(_: Request, exc: RequestValidationError) -> J
         error=ErrorDetail(
             code="VALIDATION_ERROR",
             message="Request validation failed.",
-            details={"errors": exc.errors()},
+            # Pydantic ctx can contain exception objects; input may contain API keys.
+            details={"errors": [
+                {key: error[key] for key in ("type", "loc", "msg") if key in error}
+                for error in exc.errors()
+            ]},
         )
     )
     return JSONResponse(status_code=422, content=jsonable_encoder(body))

backend/app/providers/anthropic_messages.py

@@ -0,0 +1,163 @@
+"""Native Anthropic Messages protocol with incrementally decoded content blocks."""
+
+import json
+from contextlib import aclosing
+
+from app.contracts import MessageRole, ModelEventType, ModelRequest
+from app.providers.base import ProviderError, ProviderToolCall, ProviderTurn
+from app.providers.http_base import (
+    UsageTracker, check_error, decode_tool_arguments, invalid_response, list_value,
+    object_value, string_value, token_count, truncated_stream,
+)
+from app.providers.openai_compatible import OpenAICompatibleProvider
+from app.providers.tool_names import mapped_tool_names
+
+
+class AnthropicMessagesProvider(OpenAICompatibleProvider):
+    stream_path = "/messages"
+
+    def _headers(self) -> dict[str, str]:
+        headers = super()._headers()
+        authorization = headers.pop("Authorization", None)
+        if authorization:
+            headers["x-api-key"] = authorization.removeprefix("Bearer ")
+        headers["anthropic-version"] = "2023-06-01"
+        return headers
+
+    def _payload(self, request: ModelRequest, *, stream: bool) -> dict[str, object]:
+        systems = [request.system] if request.system else []
+        messages = []
+        for message in request.messages:
+            if message.role == MessageRole.system:
+                systems.append(message.content)
+                continue
+            if message.role == MessageRole.tool:
+                if not message.tool_call_id:
+                    raise ProviderError("PROVIDER_INVALID_REQUEST", "Tool result requires a call identifier.")
+                role = "user"
+                content = [{"type": "tool_result", "tool_use_id": message.tool_call_id, "content": message.content}]
+            else:
+                role = message.role.value
+                content = [{"type": "text", "text": message.content}] if message.content else []
+                content += [{"type": "tool_use", "id": call.tool_call_id, "name": call.name,
+                             "input": call.arguments} for call in message.tool_calls]
+            if not content:
+                continue
+            if messages and messages[-1]["role"] == role:
+                messages[-1]["content"].extend(content)
+            else:
+                messages.append({"role": role, "content": content})
+        payload: dict[str, object] = {"model": request.model, "messages": messages,
+                                     "max_tokens": request.max_tokens or 4096, "stream": stream}
+        if systems:
+            payload["system"] = "\n\n".join(systems)
+        if request.tools:
+            payload["tools"] = [{"name": tool.name, "description": tool.description,
+                                 "input_schema": tool.parameters} for tool in request.tools]
+        if request.temperature is not None:
+            payload["temperature"] = request.temperature
+        if request.response_format is not None:
+            format_ = request.response_format
+            if format_.get("type") != "json_schema":
+                raise ProviderError("PROVIDER_INVALID_REQUEST", "Messages requires a JSON schema response format.")
+            schema = object_value(format_.get("json_schema"))
+            payload["output_config"] = {"format": {"type": "json_schema", "schema": object_value(schema.get("schema"))}}
+        return payload
+
+    @mapped_tool_names
+    async def complete(self, request: ModelRequest) -> ProviderTurn:
+        data = await self._request("POST", self.stream_path, json=self._payload(request, stream=False))
+        texts = []
+        calls = []
+        for raw in list_value(data.get("content")):
+            block = object_value(raw)
+            if block.get("type") == "text":
+                texts.append(string_value(block.get("text")))
+            elif block.get("type") == "tool_use":
+                calls.append(ProviderToolCall(
+                    tool_call_id=string_value(block.get("id"), nonempty=True),
+                    name=string_value(block.get("name"), nonempty=True),
+                    arguments=decode_tool_arguments(block.get("input")),
+                ))
+        return ProviderTurn(text="".join(texts) or None, tool_calls=calls,
+                            **UsageTracker(cache_tokens=True).update(data.get("usage") or {}))
+
+    async def _events(self, request: ModelRequest):
+        blocks: dict[int, dict] = {}
+        usage = UsageTracker(cache_tokens=True)
+        started = False
+        async with aclosing(self._stream_json(self._payload(request, stream=True))) as chunks:
+            async for data in chunks:
+                kind = string_value(data.get("type"), nonempty=True)
+                if kind == "message_start":
+                    if started:
+                        raise invalid_response()
+                    started = True
+                    message = object_value(data.get("message"))
+                    check_error(message)
+                    if message.get("usage") is not None:
+                        yield ModelEventType.usage, usage.update(message["usage"])
+                elif kind == "content_block_start":
+                    index = token_count(data.get("index"))
+                    if not started or index in blocks:
+                        raise invalid_response()
+                    block = dict(object_value(data.get("content_block")))
+                    blocks[index] = block
+                    block["closed"] = False
+                    if block.get("type") == "tool_use":
+                        block["id"] = string_value(block.get("id"), nonempty=True)
+                        block["name"] = string_value(block.get("name"), nonempty=True)
+                        block["arguments"] = ""
+                        block["input"] = object_value(block.get("input", {}))
+                        yield ModelEventType.tool_call_start, {"tool_call_id": block["id"], "name": block["name"]}
+                    elif block.get("type") == "text" and block.get("text"):
+                        yield ModelEventType.text_delta, {"text": string_value(block["text"])}
+                    elif block.get("type") == "thinking" and block.get("thinking"):
+                        yield ModelEventType.thinking_delta, {"text": string_value(block["thinking"])}
+                elif kind == "content_block_delta":
+                    block = blocks.get(token_count(data.get("index")))
+                    if block is None or block["closed"]:
+                        raise invalid_response()
+                    delta = object_value(data.get("delta"))
+                    delta_type = delta.get("type")
+                    if delta_type == "text_delta":
+                        if block.get("type") != "text":
+                            raise invalid_response()
+                        yield ModelEventType.text_delta, {"text": string_value(delta.get("text"))}
+                    elif delta_type == "thinking_delta":
+                        if block.get("type") != "thinking":
+                            raise invalid_response()
+                        yield ModelEventType.thinking_delta, {"text": string_value(delta.get("thinking"))}
+                    elif delta_type == "input_json_delta" and block.get("type") == "tool_use":
+                        fragment = string_value(delta.get("partial_json"))
+                        block["arguments"] += fragment
+                        yield ModelEventType.tool_call_delta, {"tool_call_id": block["id"], "arguments_delta": fragment}
+                    # Signatures and future delta types have no representation in ModelEvent.
+                elif kind == "content_block_stop":
+                    block = blocks.get(token_count(data.get("index")))
+                    if block is None or block["closed"]:
+                        raise invalid_response()
+                    block["closed"] = True
+                    if block.get("type") == "tool_use":
+                        if block["arguments"]:
+                            decode_tool_arguments(block["arguments"])
+                        else:
+                            yield ModelEventType.tool_call_delta, {
+                                "tool_call_id": block["id"], "arguments_delta": json.dumps(block["input"]),
+                            }
+                        yield ModelEventType.tool_call_end, {"tool_call_id": block["id"]}
+                elif kind == "message_delta":
+                    if not started:
+                        raise invalid_response()
+                    object_value(data.get("delta"))
+                    if data.get("usage") is not None:
+                        yield ModelEventType.usage, usage.update(data["usage"])
+                elif kind == "message_stop":
+                    if not started:
+                        raise invalid_response()
+                    if any(not block["closed"] for block in blocks.values()):
+                        raise truncated_stream()
+                    return
+                elif kind == "[DONE]":
+                    raise truncated_stream()
+        raise truncated_stream()

backend/app/providers/factory.py

@@ -16,6 +16,18 @@ class ProviderFactory:
         self.credentials = ProviderCredentialResolver(credentials)
 
     def build(self, config: ProviderConfig) -> ModelProvider:
+        if config.provider_type == ProviderType.openai_responses:
+            from app.providers.openai_responses import OpenAIResponsesProvider
+            return OpenAIResponsesProvider(
+                base_url=config.base_url or "https://api.openai.com/v1",
+                credential_id=config.credential_id, credentials=self.credentials,
+            )
+        if config.provider_type == ProviderType.anthropic_messages:
+            from app.providers.anthropic_messages import AnthropicMessagesProvider
+            return AnthropicMessagesProvider(
+                base_url=config.base_url or "https://api.anthropic.com/v1",
+                credential_id=config.credential_id, credentials=self.credentials,
+            )
         if config.provider_type in {
             ProviderType.openai_chat,
             ProviderType.openai_compatible,
@@ -31,7 +43,7 @@ class ProviderFactory:
 
     @staticmethod
     def presets() -> list[ProviderPreset]:
-        return [
+        presets = [
             ProviderPreset(
                 preset_id="openai",
                 name="OpenAI",
@@ -54,12 +66,45 @@ class ProviderFactory:
                 requires_credential=False,
             ),
         ]
+        # General API endpoints. Coding-plan endpoints and keys are separate products.
+        domestic = [
+            ("kimi", "Kimi / 月之暗面", "https://api.moonshot.cn/v1", [], "长上下文对话；模型以账号权限为准。"),
+            ("qwen", "阿里云百炼", "https://dashscope.aliyuncs.com/compatible-mode/v1", [ModelCapability.embedding], "中国内地兼容接口；海外地域需修改地址。"),
+            ("zhipu", "智谱 GLM", "https://open.bigmodel.cn/api/paas/v4", [ModelCapability.embedding], "通用 API；Coding Plan 请使用其专用地址。"),
+            ("volcengine", "火山方舟 / 豆包", "https://ark.cn-beijing.volces.com/api/v3", [ModelCapability.embedding], "按账号填写模型 ID 或推理接入点 ID。"),
+            ("siliconflow", "硅基流动", "https://api.siliconflow.cn/v1", [ModelCapability.embedding, ModelCapability.transcription], "支持兼容 Embedding 和音频转写接口。"),
+            ("baidu", "百度千帆", "https://qianfan.baidubce.com/v2", [ModelCapability.embedding], "使用千帆 API Key；模型列表取决于账号。"),
+            ("hunyuan", "腾讯混元", "https://api.hunyuan.cloud.tencent.com/v1", [], "OpenAI 兼容对话接口。"),
+            ("minimax", "MiniMax", "https://api.minimaxi.com/v1", [], "文本对话兼容接口；其他媒体协议需独立适配。"),
+            ("stepfun", "阶跃星辰", "https://api.stepfun.com/v1", [], "通用 API；Step Plan 请使用其专用地址。"),
+        ]
+        for preset_id, name, url, extra, description in domestic:
+            presets.append(ProviderPreset(
+                preset_id=preset_id, name=name, provider_type=ProviderType.openai_compatible,
+                base_url=url, default_credential_id=preset_id, logo_id=preset_id,
+                capabilities=[ModelCapability.chat, *extra], description=description,
+            ))
+        presets.extend([
+            ProviderPreset(preset_id="openai-responses", name="OpenAI Responses", provider_type=ProviderType.openai_responses,
+                           base_url="https://api.openai.com/v1", default_credential_id="openai", logo_id="openai"),
+            ProviderPreset(preset_id="anthropic", name="Anthropic / Claude", provider_type=ProviderType.anthropic_messages,
+                           base_url="https://api.anthropic.com/v1", default_credential_id="anthropic", logo_id="anthropic"),
+        ])
+        for preset in presets:
+            if preset.logo_id == "custom":
+                preset.logo_id = preset.preset_id
+            if not preset.capabilities:
+                preset.capabilities = [ModelCapability.chat]
+        presets[0].capabilities += [ModelCapability.embedding, ModelCapability.transcription]
+        return presets
 
     @staticmethod
     def capabilities(provider_type: ProviderType) -> list[ModelCapability]:
         if provider_type in {
             ProviderType.openai_chat,
             ProviderType.openai_compatible,
+            ProviderType.openai_responses,
+            ProviderType.anthropic_messages,
         }:
             return [
                 ModelCapability.chat,

backend/app/providers/http_base.py

@@ -1,9 +1,13 @@
 import json
 from collections.abc import AsyncIterator
+from contextlib import aclosing
 from datetime import datetime, timezone
 
+import httpx
+
 from app.contracts import ModelEvent, ModelEventType, ModelRequest
 from app.providers.base import ProviderError, ProviderTurn
+from app.providers.tool_names import prepare_tool_names
 
 
 class TurnStreamingMixin:
@@ -80,3 +84,229 @@ def decode_tool_arguments(value: object) -> dict[str, object]:
     if not isinstance(decoded, dict):
         raise ProviderError("PROVIDER_INVALID_RESPONSE", "Tool arguments must be an object.")
     return decoded
+
+
+def invalid_response() -> ProviderError:
+    return ProviderError("PROVIDER_INVALID_RESPONSE", "Provider returned an invalid response.")
+
+
+def truncated_stream() -> ProviderError:
+    return ProviderError("PROVIDER_STREAM_TRUNCATED", "Provider stream ended before completion.")
+
+
+def object_value(value: object) -> dict:
+    if not isinstance(value, dict):
+        raise invalid_response()
+    return value
+
+
+def list_value(value: object) -> list:
+    if not isinstance(value, list):
+        raise invalid_response()
+    return value
+
+
+def string_value(value: object, *, nonempty: bool = False) -> str:
+    if not isinstance(value, str) or (nonempty and not value):
+        raise invalid_response()
+    return value
+
+
+def token_count(value: object) -> int:
+    if isinstance(value, bool) or not isinstance(value, int) or value < 0:
+        raise invalid_response()
+    return value
+
+
+def remote_error(value: object) -> ProviderError:
+    # Never reflect upstream messages, URLs, request bodies or credentials.
+    error = value if isinstance(value, dict) else {}
+    code = error.get("code") or error.get("type")
+    mapping = {
+        "authentication_error": "PROVIDER_AUTH_FAILED",
+        "invalid_api_key": "PROVIDER_AUTH_FAILED",
+        "permission_error": "PROVIDER_AUTH_FAILED",
+        "rate_limit_error": "PROVIDER_RATE_LIMITED",
+        "rate_limit_exceeded": "PROVIDER_RATE_LIMITED",
+        "insufficient_quota": "PROVIDER_RATE_LIMITED",
+        "not_found_error": "MODEL_NOT_FOUND",
+        "model_not_found": "MODEL_NOT_FOUND",
+        "invalid_request_error": "PROVIDER_INVALID_REQUEST",
+        "context_length_exceeded": "PROVIDER_INVALID_REQUEST",
+    }
+    mapped = mapping.get(code, "PROVIDER_UNAVAILABLE") if isinstance(code, str) else "PROVIDER_UNAVAILABLE"
+    return ProviderError(mapped, "Provider could not complete the request.")
+
+
+def check_error(data: dict) -> None:
+    if data.get("error") is not None or data.get("type") == "error":
+        raise remote_error(data.get("error") or data)
+
+
+class UsageTracker:
+    """Merge cumulative snapshots, including partial usage updates."""
+
+    def __init__(self, input_key: str = "input_tokens", output_key: str = "output_tokens",
+                 *, cache_tokens: bool = False) -> None:
+        self.input_key = input_key
+        self.output_key = output_key
+        self.cache_tokens = cache_tokens
+        self.counts: dict[str, int] = {}
+
+    def update(self, value: object) -> dict[str, int]:
+        usage = object_value(value)
+        keys = [self.input_key, self.output_key]
+        if self.cache_tokens:
+            keys += ["cache_creation_input_tokens", "cache_read_input_tokens"]
+        for key in keys:
+            if key in usage:
+                self.counts[key] = max(self.counts.get(key, 0), token_count(usage[key]))
+        inputs = self.counts.get(self.input_key, 0)
+        if self.cache_tokens:
+            inputs += sum(self.counts.get(key, 0) for key in keys[2:])
+        return {"input_tokens": inputs, "output_tokens": self.counts.get(self.output_key, 0)}
+
+
+class EventStreamingMixin:
+    async def stream(self, request: ModelRequest) -> AsyncIterator[ModelEvent]:
+        sequence = 0
+        status = "completed"
+        try:
+            request, originals = prepare_tool_names(request)
+            # Closing the public iterator must synchronously close every nested iterator.
+            async with aclosing(self._events(request)) as events:
+                async for kind, data in events:
+                    if kind == ModelEventType.tool_call_start and "name" in data:
+                        data = {**data, "name": originals.get(data["name"], data["name"])}
+                    if kind == ModelEventType.usage:
+                        data = {**data, "total_tokens": data["input_tokens"] + data["output_tokens"]}
+                    yield ModelEvent(event=kind, data=data, sequence=sequence,
+                                     timestamp=datetime.now(timezone.utc))
+                    sequence += 1
+        except ProviderError as exc:
+            status = "failed"
+            yield ModelEvent(event=ModelEventType.error, sequence=sequence,
+                             data={"code": exc.code, "message": exc.message},
+                             timestamp=datetime.now(timezone.utc))
+            sequence += 1
+        except (ValueError, TypeError, KeyError, IndexError, AttributeError, OverflowError):
+            status = "failed"
+            error = invalid_response()
+            yield ModelEvent(event=ModelEventType.error, sequence=sequence,
+                             data={"code": error.code, "message": error.message},
+                             timestamp=datetime.now(timezone.utc))
+            sequence += 1
+        # CancelledError and GeneratorExit deliberately propagate without a Done event.
+        yield ModelEvent(event=ModelEventType.done, sequence=sequence,
+                         data={"status": status},
+                         timestamp=datetime.now(timezone.utc))
+
+
+async def sse_objects(response: httpx.Response) -> AsyncIterator[dict]:
+    """Read SSE frames, accepting the adjacent data lines used by some gateways."""
+    parts: list[str] = []
+    event_name = ""
+
+    def decode() -> dict:
+        value = "\n".join(parts)
+        if value.strip() == "[DONE]":
+            return {"type": "[DONE]"}
+        try:
+            data = object_value(json.loads(value))
+        except (ValueError, TypeError) as exc:
+            raise invalid_response() from exc
+        if event_name and "type" not in data:
+            data["type"] = event_name
+        check_error(data)
+        return data
+
+    async for line in response.aiter_lines():
+        if not line:
+            if parts:
+                yield decode()
+            parts = []
+            event_name = ""
+        elif line.startswith(":"):
+            continue
+        elif line.startswith("event:"):
+            if parts:
+                yield decode()
+                parts = []
+            event_name = line[6:].strip()
+        elif line.startswith("data:"):
+            if parts:
+                # Legacy compatible endpoints sometimes omit blank separators.
+                try:
+                    json.loads("\n".join(parts))
+                except ValueError:
+                    pass
+                else:
+                    yield decode()
+                    parts = []
+                    event_name = ""
+            parts.append(line[5:].removeprefix(" "))
+    if parts:
+        yield decode()
+
+
+class HTTPProviderMixin:
+    stream_path = "/chat/completions"
+    stream_format = "sse"
+
+    def _headers(self) -> dict[str, str]:
+        return {"Content-Type": "application/json"}
+
+    @staticmethod
+    def _status_error(exc: httpx.HTTPStatusError) -> ProviderError:
+        status = exc.response.status_code
+        code = {400: "PROVIDER_INVALID_REQUEST", 401: "PROVIDER_AUTH_FAILED",
+                403: "PROVIDER_AUTH_FAILED", 404: "MODEL_NOT_FOUND",
+                408: "PROVIDER_TIMEOUT", 413: "PROVIDER_INVALID_REQUEST",
+                422: "PROVIDER_INVALID_REQUEST", 429: "PROVIDER_RATE_LIMITED"}.get(
+                    status, "PROVIDER_UNAVAILABLE")
+        return ProviderError(code, f"Provider returned HTTP {status}.")
+
+    async def _request(self, method: str, path: str, **kwargs) -> dict:
+        headers = self._headers()
+        try:
+            async with httpx.AsyncClient(timeout=self.timeout_seconds, transport=self.transport) as client:
+                response = await client.request(method, f"{self.base_url}{path}", headers=headers, **kwargs)
+                response.raise_for_status()
+                data = object_value(response.json())
+                check_error(data)
+                return data
+        except httpx.TimeoutException as exc:
+            raise ProviderError("PROVIDER_TIMEOUT", "Provider request timed out.") from exc
+        except httpx.HTTPStatusError as exc:
+            raise self._status_error(exc) from exc
+        except httpx.HTTPError as exc:
+            raise ProviderError("PROVIDER_UNAVAILABLE", "Provider is unavailable.") from exc
+        except (ValueError, TypeError) as exc:
+            raise invalid_response() from exc
+
+    async def _stream_json(self, payload: dict[str, object]) -> AsyncIterator[dict]:
+        headers = self._headers()
+        headers["Accept"] = "text/event-stream" if self.stream_format == "sse" else "application/x-ndjson"
+        try:
+            async with httpx.AsyncClient(timeout=self.timeout_seconds, transport=self.transport) as client:
+                async with client.stream("POST", f"{self.base_url}{self.stream_path}",
+                                         headers=headers, json=payload) as response:
+                    response.raise_for_status()
+                    if self.stream_format == "sse":
+                        async with aclosing(sse_objects(response)) as objects:
+                            async for data in objects:
+                                yield data
+                    else:
+                        async for line in response.aiter_lines():
+                            if line.strip():
+                                data = object_value(json.loads(line))
+                                check_error(data)
+                                yield data
+        except httpx.TimeoutException as exc:
+            raise ProviderError("PROVIDER_TIMEOUT", "Provider request timed out.") from exc
+        except httpx.HTTPStatusError as exc:
+            raise self._status_error(exc) from exc
+        except httpx.HTTPError as exc:
+            raise ProviderError("PROVIDER_UNAVAILABLE", "Provider is unavailable.") from exc
+        except (ValueError, TypeError) as exc:
+            raise invalid_response() from exc

backend/app/providers/ollama.py

@@ -1,16 +1,22 @@
-from uuid import uuid4
 import json
-from collections.abc import AsyncIterator
+from contextlib import aclosing
-from datetime import datetime, timezone
+from uuid import uuid4
 
 import httpx
 
-from app.contracts import ModelCapability, ModelEvent, ModelEventType, ModelInfo, ModelRequest
+from app.contracts import MessageRole, ModelCapability, ModelEventType, ModelInfo, ModelRequest
 from app.providers.base import ProviderError, ProviderToolCall, ProviderTurn
-from app.providers.http_base import TurnStreamingMixin, decode_tool_arguments
+from app.providers.tool_names import mapped_tool_names
+from app.providers.http_base import (
+    EventStreamingMixin, HTTPProviderMixin, UsageTracker, decode_tool_arguments,
+    invalid_response, list_value, object_value, string_value, truncated_stream,
+)
 
 
-class OllamaProvider(TurnStreamingMixin):
+class OllamaProvider(EventStreamingMixin, HTTPProviderMixin):
+    stream_path = "/api/chat"
+    stream_format = "jsonl"
+
     def __init__(
         self,
         base_url: str = "http://127.0.0.1:11434",
@@ -21,126 +27,55 @@ class OllamaProvider(TurnStreamingMixin):
         self.timeout_seconds = timeout_seconds
         self.transport = transport
 
+    @mapped_tool_names
     async def complete(self, request: ModelRequest) -> ProviderTurn:
-        messages = []
+        data = await self._request("POST", self.stream_path, json=self._chat_payload(request, stream=False))
-        if request.system:
+        message = object_value(data.get("message"))
-            messages.append({"role": "system", "content": request.system})
+        calls = [self._tool_call(raw) for raw in list_value(message.get("tool_calls", []))]
-        for message in request.messages:
+        content = message.get("content")
-            item: dict[str, object] = {
+        if content is not None:
-                "role": message.role.value,
+            content = string_value(content)
-                "content": message.content,
+        return ProviderTurn(text=content or None, tool_calls=calls,
-            }
+                            **UsageTracker("prompt_eval_count", "eval_count").update(data))
-            if message.tool_calls:
+
-                item["tool_calls"] = [
+    @staticmethod
-                    {
+    def _tool_call(raw: object) -> ProviderToolCall:
-                        "function": {
+        call = object_value(raw)
-                            "name": call.name,
+        function = object_value(call.get("function"))
-                            "arguments": call.arguments,
+        return ProviderToolCall(
-                        }
+            tool_call_id=string_value(call.get("id") or f"call_{uuid4().hex}"),
-                    }
+            name=string_value(function.get("name"), nonempty=True),
-                    for call in message.tool_calls
+            arguments=decode_tool_arguments(function.get("arguments", {})),
-                ]
-            messages.append(item)
-        payload: dict[str, object] = {
-            "model": request.model,
-            "messages": messages,
-            "stream": False,
-        }
-        if request.tools:
-            payload["tools"] = [
-                {
-                    "type": "function",
-                    "function": {
-                        "name": tool.name,
-                        "description": tool.description,
-                        "parameters": tool.parameters,
-                    },
-                }
-                for tool in request.tools
-            ]
-        data = await self._request("POST", "/api/chat", json=payload)
-        message = data.get("message") or {}
-        tool_calls = []
-        for raw_call in message.get("tool_calls") or []:
-            function = raw_call.get("function") or {}
-            tool_calls.append(
-                ProviderToolCall(
-                    tool_call_id=raw_call.get("id") or f"call_{uuid4().hex}",
-                    name=function.get("name") or "",
-                    arguments=decode_tool_arguments(function.get("arguments", {})),
-                )
-            )
-        return ProviderTurn(
-            text=message.get("content") or None,
-            tool_calls=tool_calls,
-            input_tokens=int(data.get("prompt_eval_count") or 0),
-            output_tokens=int(data.get("eval_count") or 0),
         )
 
-    async def list_models(self) -> list[ModelInfo]:
+    async def _events(self, request: ModelRequest):
-        data = await self._request("GET", "/api/tags")
+        usage = UsageTracker("prompt_eval_count", "eval_count")
-        return [
+        async with aclosing(self._stream_json(self._chat_payload(request, stream=True))) as chunks:
-            ModelInfo(
+            async for data in chunks:
-                model=item["name"],
+                message = object_value(data.get("message", {}))
-                display_name=item.get("name", ""),
-                capabilities=[ModelCapability.chat, ModelCapability.streaming],
-            )
-            for item in data.get("models", [])
-            if isinstance(item, dict) and item.get("name")
-        ]
-
-    async def stream(self, request: ModelRequest) -> AsyncIterator[ModelEvent]:
-        payload = self._chat_payload(request, stream=True)
-        sequence = 0
-
-        def event(kind: ModelEventType, data: dict | None = None) -> ModelEvent:
-            nonlocal sequence
-            item = ModelEvent(
-                event=kind, sequence=sequence, data=data or {},
-                timestamp=datetime.now(timezone.utc),
-            )
-            sequence += 1
-            return item
-
-        try:
-            async for data in self._stream_json(payload):
-                message = data.get("message") or {}
                 if message.get("thinking"):
-                    yield event(ModelEventType.thinking_delta, {"text": message["thinking"]})
+                    yield ModelEventType.thinking_delta, {"text": string_value(message["thinking"])}
                 if message.get("content"):
-                    yield event(ModelEventType.text_delta, {"text": message["content"]})
+                    yield ModelEventType.text_delta, {"text": string_value(message["content"])}
-                for raw_call in message.get("tool_calls") or []:
+                for raw in list_value(message.get("tool_calls", [])):
-                    function = raw_call.get("function") or {}
+                    call = self._tool_call(raw)
-                    call_id = raw_call.get("id") or f"call_{uuid4().hex}"
+                    yield ModelEventType.tool_call_start, {"tool_call_id": call.tool_call_id, "name": call.name}
-                    yield event(
+                    yield ModelEventType.tool_call_delta, {
-                        ModelEventType.tool_call_start,
+                        "tool_call_id": call.tool_call_id,
-                        {"tool_call_id": call_id, "name": function.get("name") or ""},
+                        "arguments_delta": json.dumps(call.arguments, ensure_ascii=False),
-                    )
+                    }
-                    yield event(
+                    yield ModelEventType.tool_call_end, {"tool_call_id": call.tool_call_id}
-                        ModelEventType.tool_call_delta,
+                if "done" in data and not isinstance(data["done"], bool):
-                        {
+                    raise invalid_response()
-                            "tool_call_id": call_id,
+                if "prompt_eval_count" in data or "eval_count" in data or data.get("done"):
-                            "arguments_delta": json.dumps(
+                    yield ModelEventType.usage, usage.update(data)
-                                function.get("arguments") or {}, ensure_ascii=False
+                if data.get("done") is True:
-                            ),
+                    return
-                        },
+        raise truncated_stream()
-                    )
-                    yield event(ModelEventType.tool_call_end, {"tool_call_id": call_id})
-                if data.get("done"):
-                    yield event(
-                        ModelEventType.usage,
-                        {
-                            "input_tokens": int(data.get("prompt_eval_count") or 0),
-                            "output_tokens": int(data.get("eval_count") or 0),
-                        },
-                    )
-            yield event(ModelEventType.done)
-        except ProviderError as exc:
-            yield event(ModelEventType.error, {"code": exc.code, "message": exc.message})
-            yield event(ModelEventType.done)
 
     def _chat_payload(self, request: ModelRequest, *, stream: bool) -> dict[str, object]:
         messages = []
+        names: dict[str, str] = {}
         if request.system:
             messages.append({"role": "system", "content": request.system})
         for message in request.messages:
@@ -150,53 +85,49 @@ class OllamaProvider(TurnStreamingMixin):
                     {"function": {"name": call.name, "arguments": call.arguments}}
                     for call in message.tool_calls
                 ]
+                names.update({call.tool_call_id: call.name for call in message.tool_calls})
+            if message.role == MessageRole.tool:
+                name = message.name or names.get(message.tool_call_id or "")
+                if name:
+                    item["tool_name"] = name
             messages.append(item)
         payload: dict[str, object] = {
-            "model": request.model, "messages": messages, "stream": stream
+            "model": request.model, "messages": messages, "stream": stream,
         }
         if request.tools:
             payload["tools"] = [
-                {
+                {"type": "function", "function": {
-                    "type": "function",
+                    "name": tool.name, "description": tool.description, "parameters": tool.parameters,
-                    "function": {
+                }} for tool in request.tools
-                        "name": tool.name,
-                        "description": tool.description,
-                        "parameters": tool.parameters,
-                    },
-                }
-                for tool in request.tools
             ]
+        options = {}
+        if request.temperature is not None:
+            options["temperature"] = request.temperature
+        if request.max_tokens is not None:
+            options["num_predict"] = request.max_tokens
+        if options:
+            payload["options"] = options
+        if request.response_format:
+            format_ = request.response_format
+            if format_.get("type") == "json_object":
+                payload["format"] = "json"
+            elif format_.get("type") == "json_schema":
+                payload["format"] = object_value(object_value(format_.get("json_schema")).get("schema"))
+            else:
+                payload["format"] = format_
         return payload
 
-    async def _stream_json(self, payload: dict[str, object]) -> AsyncIterator[dict]:
+    async def list_models(self) -> list[ModelInfo]:
-        try:
+        data = await self._request("GET", "/api/tags")
-            async with httpx.AsyncClient(
+        return [
-                timeout=self.timeout_seconds, transport=self.transport
+            ModelInfo(
-            ) as client:
+                model=string_value(item["name"]), display_name=item["name"],
-                async with client.stream(
+                capabilities=([ModelCapability.embedding] if "embed" in item["name"].lower()
-                    "POST", f"{self.base_url}/api/chat", json=payload
+                              else [ModelCapability.chat, ModelCapability.streaming]),
-                ) as response:
+            )
-                    response.raise_for_status()
+            for item in list_value(data.get("models"))
-                    async for line in response.aiter_lines():
+            if isinstance(item, dict) and isinstance(item.get("name"), str) and item["name"]
-                        if not line.strip():
+        ]
-                            continue
-                        try:
-                            data = json.loads(line)
-                        except json.JSONDecodeError as exc:
-                            raise ProviderError(
-                                "PROVIDER_INVALID_RESPONSE", "Ollama returned invalid JSONL."
-                            ) from exc
-                        if isinstance(data, dict):
-                            yield data
-        except httpx.TimeoutException as exc:
-            raise ProviderError("PROVIDER_TIMEOUT", "Ollama request timed out.") from exc
-        except httpx.HTTPStatusError as exc:
-            raise ProviderError(
-                "MODEL_NOT_FOUND" if exc.response.status_code == 404 else "PROVIDER_UNAVAILABLE",
-                f"Ollama returned HTTP {exc.response.status_code}.",
-            ) from exc
-        except httpx.HTTPError as exc:
-            raise ProviderError("PROVIDER_UNAVAILABLE", "Ollama is unavailable.") from exc
 
     async def test_connection(self, model: str | None = None) -> tuple[bool, str]:
         try:
@@ -206,24 +137,3 @@ class OllamaProvider(TurnStreamingMixin):
         if model and model not in {item.model for item in models}:
             return False, f"Model is not installed: {model}"
         return True, f"Connected; discovered {len(models)} local model(s)."
-
-    async def _request(self, method: str, path: str, **kwargs) -> dict:
-        try:
-            async with httpx.AsyncClient(
-                timeout=self.timeout_seconds, transport=self.transport
-            ) as client:
-                response = await client.request(method, f"{self.base_url}{path}", **kwargs)
-                response.raise_for_status()
-                data = response.json()
-        except httpx.TimeoutException as exc:
-            raise ProviderError("PROVIDER_TIMEOUT", "Ollama request timed out.") from exc
-        except httpx.HTTPStatusError as exc:
-            raise ProviderError(
-                "MODEL_NOT_FOUND" if exc.response.status_code == 404 else "PROVIDER_UNAVAILABLE",
-                f"Ollama returned HTTP {exc.response.status_code}.",
-            ) from exc
-        except (httpx.HTTPError, ValueError) as exc:
-            raise ProviderError("PROVIDER_UNAVAILABLE", "Ollama is unavailable.") from exc
-        if not isinstance(data, dict):
-            raise ProviderError("PROVIDER_INVALID_RESPONSE", "Ollama returned non-object JSON.")
-        return data

backend/app/providers/openai_compatible.py

@@ -1,24 +1,20 @@
 import json
-from collections.abc import AsyncIterator
+from contextlib import aclosing
-from datetime import datetime, timezone
 from uuid import uuid4
 
 import httpx
 
-from app.contracts import (
+from app.contracts import MessageRole, ModelCapability, ModelEventType, ModelInfo, ModelRequest
-    MessageRole,
-    ModelCapability,
-    ModelEvent,
-    ModelEventType,
-    ModelInfo,
-    ModelRequest,
-)
 from app.providers.base import ProviderError, ProviderToolCall, ProviderTurn
 from app.providers.credentials import CredentialResolver, CredentialStoreError
-from app.providers.http_base import TurnStreamingMixin, decode_tool_arguments
+from app.providers.tool_names import mapped_tool_names
+from app.providers.http_base import (
+    EventStreamingMixin, HTTPProviderMixin, UsageTracker, decode_tool_arguments,
+    invalid_response, list_value, object_value, string_value, token_count, truncated_stream,
+)
 
 
-class OpenAICompatibleProvider(TurnStreamingMixin):
+class OpenAICompatibleProvider(EventStreamingMixin, HTTPProviderMixin):
     def __init__(
         self,
         base_url: str,
@@ -33,50 +29,37 @@ class OpenAICompatibleProvider(TurnStreamingMixin):
         self.timeout_seconds = timeout_seconds
         self.transport = transport
 
+    @mapped_tool_names
     async def complete(self, request: ModelRequest) -> ProviderTurn:
-        payload = self._payload(request, stream=False)
+        data = await self._request("POST", self.stream_path, json=self._payload(request, stream=False))
-
+        choices = list_value(data.get("choices"))
-        data = await self._request("POST", "/chat/completions", json=payload)
+        if not choices:
-        try:
+            raise invalid_response()
-            message = data["choices"][0]["message"]
+        message = object_value(object_value(choices[0]).get("message"))
-        except (KeyError, IndexError, TypeError) as exc:
+        calls = []
-            raise ProviderError("PROVIDER_INVALID_RESPONSE", "Missing completion message.") from exc
+        for raw in list_value(message.get("tool_calls", [])):
-
+            raw = object_value(raw)
-        tool_calls = []
+            function = object_value(raw.get("function"))
-        for raw_call in message.get("tool_calls") or []:
+            calls.append(ProviderToolCall(
-            function = raw_call.get("function") or {}
+                tool_call_id=string_value(raw.get("id") or f"call_{uuid4().hex}"),
-            tool_calls.append(
+                name=string_value(function.get("name"), nonempty=True),
-                ProviderToolCall(
+                arguments=decode_tool_arguments(function.get("arguments", "{}")),
-                    tool_call_id=raw_call.get("id") or f"call_{uuid4().hex}",
+            ))
-                    name=function.get("name") or "",
+        text = message.get("content")
-                    arguments=decode_tool_arguments(function.get("arguments", "{}")),
+        if text is not None:
-                )
+            text = string_value(text)
-            )
+        usage = UsageTracker("prompt_tokens", "completion_tokens").update(data.get("usage") or {})
-        usage = data.get("usage") or {}
+        return ProviderTurn(text=text, tool_calls=calls, **usage)
-        return ProviderTurn(
-            text=message.get("content"),
-            tool_calls=tool_calls,
-            input_tokens=int(usage.get("prompt_tokens") or 0),
-            output_tokens=int(usage.get("completion_tokens") or 0),
-        )
 
     def _payload(self, request: ModelRequest, *, stream: bool) -> dict[str, object]:
         payload: dict[str, object] = {
-            "model": request.model,
+            "model": request.model, "messages": self._messages(request), "stream": stream,
-            "messages": self._messages(request),
-            "stream": stream,
         }
         if request.tools:
             payload["tools"] = [
-                {
+                {"type": "function", "function": {
-                    "type": "function",
+                    "name": tool.name, "description": tool.description, "parameters": tool.parameters,
-                    "function": {
+                }} for tool in request.tools
-                        "name": tool.name,
-                        "description": tool.description,
-                        "parameters": tool.parameters,
-                    },
-                }
-                for tool in request.tools
             ]
         if request.temperature is not None:
             payload["temperature"] = request.temperature
@@ -84,124 +67,78 @@ class OpenAICompatibleProvider(TurnStreamingMixin):
             payload["max_tokens"] = request.max_tokens
         if request.response_format is not None:
             payload["response_format"] = request.response_format
-
+        if stream:
+            payload["stream_options"] = {"include_usage": True}
         return payload
 
-    async def stream(self, request: ModelRequest) -> AsyncIterator[ModelEvent]:
+    async def _events(self, request: ModelRequest):
-        sequence = 0
+        calls: dict[int, dict] = {}
-        open_calls: dict[int, str] = {}
+        usage = UsageTracker("prompt_tokens", "completion_tokens")
-
+        finished = False
-        def event(kind: ModelEventType, data: dict | None = None) -> ModelEvent:
+        seen = False
-            nonlocal sequence
+        async with aclosing(self._stream_json(self._payload(request, stream=True))) as chunks:
-            item = ModelEvent(
+            async for data in chunks:
-                event=kind,
+                if data.get("type") == "[DONE]":
-                sequence=sequence,
+                    if not seen:
-                data=data or {},
+                        raise invalid_response()
-                timestamp=datetime.now(timezone.utc),
+                    finished = True
-            )
+                    break
-            sequence += 1
+                if data.get("usage") is not None:
-            return item
+                    yield ModelEventType.usage, usage.update(data["usage"])
-
+                choices = list_value(data.get("choices", []))
-        try:
-            async for data in self._stream_json(self._payload(request, stream=True)):
-                usage = data.get("usage") or {}
-                if usage:
-                    yield event(
-                        ModelEventType.usage,
-                        {
-                            "input_tokens": int(usage.get("prompt_tokens") or 0),
-                            "output_tokens": int(usage.get("completion_tokens") or 0),
-                        },
-                    )
-                choices = data.get("choices") or []
                 if not choices:
                     continue
-                choice = choices[0]
+                seen = True
-                delta = choice.get("delta") or {}
+                choice = object_value(choices[0])
+                delta = object_value(choice.get("delta") or {})
                 if delta.get("reasoning_content"):
-                    yield event(
+                    yield ModelEventType.thinking_delta, {"text": string_value(delta["reasoning_content"])}
-                        ModelEventType.thinking_delta,
-                        {"text": delta["reasoning_content"]},
-                    )
                 if delta.get("content"):
-                    yield event(ModelEventType.text_delta, {"text": delta["content"]})
+                    yield ModelEventType.text_delta, {"text": string_value(delta["content"])}
-                for raw_call in delta.get("tool_calls") or []:
+                for raw in list_value(delta.get("tool_calls", [])):
-                    index = int(raw_call.get("index") or 0)
+                    raw = object_value(raw)
-                    function = raw_call.get("function") or {}
+                    index = token_count(raw.get("index", 0))
-                    call_id = raw_call.get("id") or open_calls.get(index) or f"call_{uuid4().hex}"
+                    function = object_value(raw.get("function") or {})
-                    if index not in open_calls:
+                    call = calls.setdefault(index, {"id": "", "name": "", "arguments": ""})
-                        open_calls[index] = call_id
+                    if raw.get("id"):
-                        yield event(
+                        call["id"] = string_value(raw["id"])
-                            ModelEventType.tool_call_start,
+                    if function.get("name"):
-                            {"tool_call_id": call_id, "name": function.get("name") or ""},
+                        call["name"] += string_value(function["name"])
-                        )
+                    fragment = string_value(function.get("arguments", ""))
-                    if function.get("arguments"):
+                    call["arguments"] += fragment
-                        yield event(
+                if choice.get("finish_reason"):
-                            ModelEventType.tool_call_delta,
+                    finished = True
-                            {
+        if not finished:
-                                "tool_call_id": open_calls[index],
+            raise truncated_stream()
-                                "arguments_delta": function["arguments"],
+        for call in calls.values():
-                            },
+            if not call["name"]:
-                        )
+                raise invalid_response()
-                if choice.get("finish_reason") == "tool_calls":
+            decode_tool_arguments(call["arguments"] or "{}")
-                    for call_id in open_calls.values():
+            # A name can span multiple chunks; publish only the complete identity.
-                        yield event(
+            call["id"] = call["id"] or f"call_{uuid4().hex}"
-                            ModelEventType.tool_call_end, {"tool_call_id": call_id}
+            yield ModelEventType.tool_call_start, {"tool_call_id": call["id"], "name": call["name"]}
-                        )
+            yield ModelEventType.tool_call_delta, {"tool_call_id": call["id"], "arguments_delta": call["arguments"] or "{}"}
-                    open_calls.clear()
+            yield ModelEventType.tool_call_end, {"tool_call_id": call["id"]}
-            for call_id in open_calls.values():
-                yield event(ModelEventType.tool_call_end, {"tool_call_id": call_id})
-            yield event(ModelEventType.done)
-        except ProviderError as exc:
-            yield event(ModelEventType.error, {"code": exc.code, "message": exc.message})
-            yield event(ModelEventType.done)
-
-    async def _stream_json(self, payload: dict[str, object]) -> AsyncIterator[dict]:
-        headers = self._headers()
-        try:
-            async with httpx.AsyncClient(
-                timeout=self.timeout_seconds, transport=self.transport
-            ) as client:
-                async with client.stream(
-                    "POST", f"{self.base_url}/chat/completions", headers=headers, json=payload
-                ) as response:
-                    response.raise_for_status()
-                    async for line in response.aiter_lines():
-                        if not line.startswith("data:"):
-                            continue
-                        value = line[5:].strip()
-                        if not value or value == "[DONE]":
-                            continue
-                        try:
-                            data = json.loads(value)
-                        except json.JSONDecodeError as exc:
-                            raise ProviderError(
-                                "PROVIDER_INVALID_RESPONSE", "Provider returned invalid SSE JSON."
-                            ) from exc
-                        if isinstance(data, dict):
-                            yield data
-        except httpx.TimeoutException as exc:
-            raise ProviderError("PROVIDER_TIMEOUT", "Provider request timed out.") from exc
-        except httpx.HTTPStatusError as exc:
-            raise self._status_error(exc) from exc
-        except httpx.HTTPError as exc:
-            raise ProviderError("PROVIDER_UNAVAILABLE", "Provider is unavailable.") from exc
 
     async def list_models(self) -> list[ModelInfo]:
         data = await self._request("GET", "/models")
-        return [
+        return [ModelInfo(model=string_value(item["id"]), display_name=item["id"],
-            ModelInfo(
+                          capabilities=self._model_capabilities(string_value(item["id"])))
-                model=item["id"],
+                for item in list_value(data.get("data"))
-                display_name=item["id"],
+                if isinstance(item, dict) and item.get("id")]
-                capabilities=[
+
-                    ModelCapability.chat,
+    @staticmethod
-                    ModelCapability.tool_calling,
+    def _model_capabilities(model: str) -> list[ModelCapability]:
-                    ModelCapability.streaming,
+        # /models does not advertise capabilities. Avoid known non-chat families;
-                ],
+        # these are discovery hints, not a guarantee of support by a gateway.
-            )
+        name = model.lower()
-            for item in data.get("data", [])
+        if "embed" in name or name.startswith(("bge-", "bge/")):
-            if isinstance(item, dict) and item.get("id")
+            return [ModelCapability.embedding]
-        ]
+        if any(marker in name for marker in (
+            "whisper", "tts", "transcri", "audio", "realtime", "dall-e", "image", "moderation", "rerank",
+        )):
+            return []
+        return [ModelCapability.chat]
 
     async def test_connection(self, model: str | None = None) -> tuple[bool, str]:
         try:
@@ -217,73 +154,30 @@ class OpenAICompatibleProvider(TurnStreamingMixin):
         if request.system:
             result.append({"role": "system", "content": request.system})
         for message in request.messages:
-            item: dict[str, object] = {
+            item: dict[str, object] = {"role": message.role.value, "content": message.content}
-                "role": message.role.value,
-                "content": message.content,
-            }
             if message.name:
                 item["name"] = message.name
             if message.role == MessageRole.tool and message.tool_call_id:
                 item["tool_call_id"] = message.tool_call_id
             if message.tool_calls:
                 item["tool_calls"] = [
-                    {
+                    {"id": call.tool_call_id, "type": "function", "function": {
-                        "id": call.tool_call_id,
+                        "name": call.name, "arguments": json.dumps(call.arguments),
-                        "type": "function",
+                    }} for call in message.tool_calls
-                        "function": {
-                            "name": call.name,
-                            "arguments": json.dumps(call.arguments),
-                        },
-                    }
-                    for call in message.tool_calls
                 ]
             result.append(item)
         return result
 
-    async def _request(self, method: str, path: str, **kwargs) -> dict:
-        headers = self._headers()
-        try:
-            async with httpx.AsyncClient(
-                timeout=self.timeout_seconds, transport=self.transport
-            ) as client:
-                response = await client.request(
-                    method, f"{self.base_url}{path}", headers=headers, **kwargs
-                )
-                response.raise_for_status()
-                data = response.json()
-        except httpx.TimeoutException as exc:
-            raise ProviderError("PROVIDER_TIMEOUT", "Provider request timed out.") from exc
-        except httpx.HTTPStatusError as exc:
-            raise self._status_error(exc) from exc
-        except (httpx.HTTPError, ValueError) as exc:
-            raise ProviderError("PROVIDER_UNAVAILABLE", "Provider is unavailable.") from exc
-        if not isinstance(data, dict):
-            raise ProviderError("PROVIDER_INVALID_RESPONSE", "Provider returned non-object JSON.")
-        return data
-
     def _headers(self) -> dict[str, str]:
         headers = {"Content-Type": "application/json"}
         try:
             api_key = self.credentials.resolve(self.credential_id)
         except CredentialStoreError as exc:
-            raise ProviderError(
+            raise ProviderError("PROVIDER_CREDENTIAL_UNAVAILABLE",
-                "PROVIDER_CREDENTIAL_UNAVAILABLE",
+                                "Credential could not be decrypted by the AI Core.") from exc
-                "Credential could not be decrypted by the AI Core.",
-            ) from exc
         if self.credential_id and not api_key:
-            raise ProviderError(
+            raise ProviderError("PROVIDER_CREDENTIAL_MISSING",
-                "PROVIDER_CREDENTIAL_MISSING",
+                                "Credential is not available in the AI Core process.")
-                f'Credential "{self.credential_id}" is not available in the AI Core process.',
-            )
         if api_key:
             headers["Authorization"] = f"Bearer {api_key}"
         return headers
-
-    @staticmethod
-    def _status_error(exc: httpx.HTTPStatusError) -> ProviderError:
-        code = {
-            401: "PROVIDER_AUTH_FAILED",
-            404: "MODEL_NOT_FOUND",
-            429: "PROVIDER_RATE_LIMITED",
-        }.get(exc.response.status_code, "PROVIDER_UNAVAILABLE")
-        return ProviderError(code, f"Provider returned HTTP {exc.response.status_code}.")

backend/app/providers/openai_responses.py

@@ -0,0 +1,168 @@
+"""Native /responses adapter; stateless history uses function_call/output items."""
+
+import json
+from contextlib import aclosing
+
+from app.contracts import MessageRole, ModelEventType, ModelRequest
+from app.providers.base import ProviderError, ProviderToolCall, ProviderTurn
+from app.providers.http_base import (
+    UsageTracker, check_error, decode_tool_arguments, invalid_response, list_value,
+    object_value, remote_error, string_value, token_count, truncated_stream,
+)
+from app.providers.openai_compatible import OpenAICompatibleProvider
+from app.providers.tool_names import mapped_tool_names
+
+
+class OpenAIResponsesProvider(OpenAICompatibleProvider):
+    stream_path = "/responses"
+
+    def _payload(self, request: ModelRequest, *, stream: bool) -> dict[str, object]:
+        inputs = []
+        for message in request.messages:
+            if message.role == MessageRole.tool:
+                if not message.tool_call_id:
+                    raise ProviderError("PROVIDER_INVALID_REQUEST", "Tool result requires a call identifier.")
+                inputs.append({"type": "function_call_output", "call_id": message.tool_call_id,
+                               "output": message.content})
+                continue
+            if message.content or not message.tool_calls:
+                inputs.append({"role": message.role.value, "content": message.content})
+            for call in message.tool_calls:
+                inputs.append({"type": "function_call", "call_id": call.tool_call_id,
+                               "name": call.name, "arguments": json.dumps(call.arguments)})
+        payload: dict[str, object] = {"model": request.model, "input": inputs, "stream": stream}
+        if request.system:
+            payload["instructions"] = request.system
+        if request.tools:
+            payload["tools"] = [{"type": "function", "name": tool.name,
+                                 "description": tool.description, "parameters": tool.parameters}
+                                for tool in request.tools]
+        if request.temperature is not None:
+            payload["temperature"] = request.temperature
+        if request.max_tokens is not None:
+            payload["max_output_tokens"] = request.max_tokens
+        if request.response_format is not None:
+            format_ = dict(request.response_format)
+            if format_.get("type") == "json_schema":
+                format_ = {"type": "json_schema", **object_value(format_.get("json_schema"))}
+            payload["text"] = {"format": format_}
+        return payload
+
+    @staticmethod
+    def _check_response(data: dict) -> None:
+        check_error(data)
+        status = data.get("status")
+        if status == "incomplete":
+            raise ProviderError("PROVIDER_INCOMPLETE_RESPONSE", "Provider response is incomplete.")
+        if status == "failed":
+            raise remote_error(data.get("error"))
+        if status is not None and status != "completed":
+            raise invalid_response()
+
+    @mapped_tool_names
+    async def complete(self, request: ModelRequest) -> ProviderTurn:
+        data = await self._request("POST", self.stream_path, json=self._payload(request, stream=False))
+        self._check_response(data)
+        texts = []
+        calls = []
+        for raw in list_value(data.get("output")):
+            item = object_value(raw)
+            if item.get("type") == "message":
+                for raw_part in list_value(item.get("content")):
+                    part = object_value(raw_part)
+                    if part.get("type") == "output_text":
+                        texts.append(string_value(part.get("text")))
+                    elif part.get("type") == "refusal":
+                        texts.append(string_value(part.get("refusal")))
+            elif item.get("type") == "function_call":
+                calls.append(ProviderToolCall(
+                    tool_call_id=string_value(item.get("call_id"), nonempty=True),
+                    name=string_value(item.get("name"), nonempty=True),
+                    arguments=decode_tool_arguments(item.get("arguments")),
+                ))
+        return ProviderTurn(text="".join(texts) or None, tool_calls=calls,
+                            **UsageTracker().update(data.get("usage") or {}))
+
+    async def _events(self, request: ModelRequest):
+        calls: dict[int, dict] = {}
+        usage = UsageTracker()
+
+        def finish_call(index: int, final: object = None):
+            call = calls[index]
+            if call["ended"]:
+                return []
+            events = []
+            if final is not None:
+                arguments = string_value(final)
+                if not arguments.startswith(call["arguments"]):
+                    raise invalid_response()
+                remainder = arguments[len(call["arguments"]):]
+                if remainder:
+                    events.append((ModelEventType.tool_call_delta,
+                                   {"tool_call_id": call["id"], "arguments_delta": remainder}))
+                call["arguments"] = arguments
+            decode_tool_arguments(call["arguments"])
+            call["ended"] = True
+            events.append((ModelEventType.tool_call_end, {"tool_call_id": call["id"]}))
+            return events
+
+        async with aclosing(self._stream_json(self._payload(request, stream=True))) as chunks:
+            async for data in chunks:
+                kind = string_value(data.get("type"), nonempty=True)
+                if kind in {"response.failed", "response.incomplete"}:
+                    response = object_value(data.get("response"))
+                    self._check_response({**response, "status": kind.split(".")[1]})
+                elif kind in {"response.output_text.delta", "response.refusal.delta"}:
+                    yield ModelEventType.text_delta, {"text": string_value(data.get("delta"))}
+                elif kind in {"response.reasoning_summary_text.delta", "response.reasoning_text.delta"}:
+                    yield ModelEventType.thinking_delta, {"text": string_value(data.get("delta"))}
+                elif kind in {"response.output_item.added", "response.output_item.done"}:
+                    item = object_value(data.get("item"))
+                    if item.get("type") != "function_call":
+                        continue
+                    index = token_count(data.get("output_index"))
+                    call_id = string_value(item.get("call_id"), nonempty=True)
+                    name = string_value(item.get("name"), nonempty=True)
+                    if index not in calls:
+                        calls[index] = {"id": call_id, "name": name, "arguments": "", "ended": False,
+                                        "item_id": item.get("id")}
+                        yield ModelEventType.tool_call_start, {"tool_call_id": call_id, "name": name}
+                    elif calls[index]["id"] != call_id or calls[index]["name"] != name:
+                        raise invalid_response()
+                    if kind == "response.output_item.done":
+                        for event in finish_call(index, item.get("arguments")):
+                            yield event
+                    elif item.get("arguments"):
+                        arguments = string_value(item["arguments"])
+                        calls[index]["arguments"] += arguments
+                        yield ModelEventType.tool_call_delta, {"tool_call_id": call_id, "arguments_delta": arguments}
+                elif kind in {"response.function_call_arguments.delta", "response.function_call_arguments.done"}:
+                    index = token_count(data.get("output_index"))
+                    call = calls.get(index)
+                    if call is None or (data.get("item_id") and call["item_id"] != data["item_id"]):
+                        raise invalid_response()
+                    if kind.endswith(".done"):
+                        for event in finish_call(index, data.get("arguments")):
+                            yield event
+                    else:
+                        if call["ended"]:
+                            raise invalid_response()
+                        fragment = string_value(data.get("delta"))
+                        call["arguments"] += fragment
+                        yield ModelEventType.tool_call_delta, {"tool_call_id": call["id"], "arguments_delta": fragment}
+                elif kind == "response.completed":
+                    response = object_value(data.get("response"))
+                    self._check_response(response)
+                    if any(not call["ended"] for call in calls.values()):
+                        raise truncated_stream()
+                    if response.get("usage") is not None:
+                        yield ModelEventType.usage, usage.update(response["usage"])
+                    return
+                elif kind == "[DONE]":
+                    raise truncated_stream()
+                elif kind in {"response.created", "response.in_progress"}:
+                    response = object_value(data.get("response"))
+                    check_error(response)
+                    if response.get("usage") is not None:
+                        yield ModelEventType.usage, usage.update(response["usage"])
+        raise truncated_stream()

backend/app/providers/registry.py

@@ -1,5 +1,10 @@
 from dataclasses import dataclass
 from time import perf_counter
+from pathlib import Path
+
+from app.config import get_settings
+from app.database.db import connect
+from app.errors import ApiError
 
 from app.contracts import ModelInfo, ProviderConfig, ProviderTestResponse
 from app.providers.base import ModelProvider
@@ -16,20 +21,64 @@ class RegisteredProvider:
 
 
 class ProviderRegistry:
-    def __init__(self) -> None:
+    def __init__(self, factory=None) -> None:
         self._providers: dict[str, RegisteredProvider] = {}
+        self._factory = factory
+        self._loaded_path: Path | None = None
+
+    def _restore(self) -> None:
+        if self._factory is None or self._loaded_path == get_settings().db_path:
+            return
+        conn = connect()
+        try:
+            conn.execute("CREATE TABLE IF NOT EXISTS provider_configs (provider_id TEXT PRIMARY KEY, config_json TEXT NOT NULL)")
+            restored = {}
+            for row in conn.execute("SELECT config_json FROM provider_configs"):
+                config = ProviderConfig.model_validate_json(row["config_json"])
+                if config.provider_id == "mock":
+                    raise ValueError("reserved provider")
+                restored[config.provider_id] = RegisteredProvider(config, self._factory.build(config))
+            if "mock" in self._providers:
+                restored["mock"] = self._providers["mock"]
+            self._providers = restored
+            self._loaded_path = get_settings().db_path
+        except (ValueError, TypeError) as exc:
+            raise ApiError(500, "PROVIDER_STORAGE_INVALID", "Saved provider configuration could not be loaded.") from exc
+        finally:
+            conn.close()
+
+    def _save(self, config: ProviderConfig) -> None:
+        if self._factory is None or config.provider_id == "mock":
+            return
+        conn = connect()
+        try:
+            conn.execute("INSERT OR REPLACE INTO provider_configs VALUES (?, ?)", (config.provider_id, config.model_dump_json()))
+        finally:
+            conn.close()
 
     def register(self, config: ProviderConfig, adapter: ModelProvider) -> None:
+        if config.provider_id != "mock":
+            self._restore()
         if config.provider_id in self._providers:
             raise ValueError(f"Provider already registered: {config.provider_id}")
+        self._save(config)
         self._providers[config.provider_id] = RegisteredProvider(config=config, adapter=adapter)
 
     def unregister(self, provider_id: str) -> None:
+        self._restore()
+        if self._factory is not None:
+            conn = connect()
+            try:
+                conn.execute("DELETE FROM provider_configs WHERE provider_id = ?", (provider_id,))
+            finally:
+                conn.close()
         self._providers.pop(provider_id, None)
 
     def replace(self, config: ProviderConfig, adapter: ModelProvider) -> None:
+        self._restore()
         if config.provider_id not in self._providers:
             raise ProviderNotFoundError(config.provider_id)
+        self._save(config)
         self._providers[config.provider_id] = RegisteredProvider(config=config, adapter=adapter)
 
     def get(self, provider_id: str) -> RegisteredProvider:
@@ -39,12 +88,14 @@ class ProviderRegistry:
         return provider
 
     def get_any(self, provider_id: str) -> RegisteredProvider:
+        self._restore()
         try:
             return self._providers[provider_id]
         except KeyError as exc:
             raise ProviderNotFoundError(provider_id) from exc
 
     def list_configs(self) -> list[ProviderConfig]:
+        self._restore()
         return [item.config.model_copy(deep=True) for item in self._providers.values()]
 
     async def list_models(self, provider_id: str) -> list[ModelInfo]:

backend/app/providers/routing.py

@@ -0,0 +1,291 @@
+"""Capability routing: validated remote results, then an explicit local backend.
+
+Phase E supplies HTTP adapters and injectable local contracts. Hash embeddings are
+still a development placeholder; speech models are installed in phase F.
+"""
+from __future__ import annotations
+
+import hashlib
+import json
+import math
+from dataclasses import dataclass
+from pathlib import Path
+from typing import Protocol
+
+import httpx
+
+from app.contracts import (
+    EmbeddingResult, LocalBackendStatus, ModelBinding, ModelRoutingConfig,
+    ModelRoutingResponse, ProviderType, SpeakerMatchResult,
+)
+from app.database.db import connect, transaction
+from app.errors import ApiError
+from app.providers.base import ProviderError
+from app.providers.credentials import CredentialResolver, CredentialStoreError
+from app.providers.registry import ProviderNotFoundError, ProviderRegistry
+from app.retrieval.embedding import EmbeddingProvider, HashEmbeddingProvider
+from app.retrieval.provenance import record_embedding
+
+CAPABILITIES = ("embedding", "transcription", "speaker_matching")
+HTTP_TYPES = {ProviderType.openai_chat, ProviderType.openai_compatible}
+MAX_MEDIA_BYTES = 25 * 1024 * 1024
+MAX_RESPONSE_BYTES = 16 * 1024 * 1024
+
+
+class LocalSpeechBackend(Protocol):
+    available: bool
+
+    async def transcribe(self, source: Path, language: str | None) -> str: ...
+
+    async def match(self, source: Path, reference: Path) -> float: ...
+
+
+class PendingSpeechBackend:
+    available = False
+
+    async def transcribe(self, source: Path, language: str | None) -> str:
+        raise ProviderError("LOCAL_MODEL_NOT_INSTALLED", "本地音频转写模型尚未安装，将在阶段 F 接入。")
+
+    async def match(self, source: Path, reference: Path) -> float:
+        raise ProviderError("LOCAL_MODEL_NOT_INSTALLED", "本地声纹模型尚未安装，将在阶段 F 接入。")
+
+
+@dataclass(frozen=True)
+class RoutedTranscript:
+    text: str
+    source: str
+    fallback_reason: str | None = None
+
+
+def invalid_response() -> ProviderError:
+    return ProviderError("PROVIDER_INVALID_RESPONSE", "Model API returned an invalid result.")
+
+
+def finite_number(value: object) -> bool:
+    if type(value) not in (int, float):
+        return False
+    try:
+        return math.isfinite(value)
+    except (OverflowError, ValueError):
+        return False
+
+
+class ModelRoutingService:
+    def __init__(self, providers: ProviderRegistry, credentials: CredentialResolver, *,
+                 local_embedding: EmbeddingProvider | None = None,
+                 local_speech: LocalSpeechBackend | None = None,
+                 transport: httpx.AsyncBaseTransport | None = None) -> None:
+        self.providers = providers
+        self.credentials = credentials
+        self.local_embedding = local_embedding or HashEmbeddingProvider()
+        self.local_speech = local_speech or PendingSpeechBackend()
+        self.transport = transport
+
+    @staticmethod
+    def _connection():
+        conn = connect()
+        conn.execute("CREATE TABLE IF NOT EXISTS model_routing (id INTEGER PRIMARY KEY CHECK(id=1), config_json TEXT NOT NULL)")
+        return conn
+
+    def configuration(self) -> ModelRoutingConfig:
+        conn = self._connection()
+        try:
+            row = conn.execute("SELECT config_json FROM model_routing WHERE id=1").fetchone()
+            return ModelRoutingConfig.model_validate_json(row[0]) if row else ModelRoutingConfig()
+        except ValueError as exc:
+            raise ApiError(500, "MODEL_ROUTING_STORAGE_INVALID", "Saved model routing could not be loaded.") from exc
+        finally:
+            conn.close()
+
+    def describe(self) -> ModelRoutingResponse:
+        return ModelRoutingResponse(config=self.configuration(), local_backends=[
+            LocalBackendStatus(capability="embedding", status="placeholder" if isinstance(self.local_embedding, HashEmbeddingProvider) else "ready",
+                               message="当前为 hash-v1 确定性占位向量，真实本地语义模型尚未集成。" if isinstance(self.local_embedding, HashEmbeddingProvider) else "本地 Embedding 模型已就绪。"),
+            *[LocalBackendStatus(capability=capability, status="ready" if self.local_speech.available else "not_installed",
+                                 message="本地模型已就绪。" if self.local_speech.available else "阶段 F 接入本地模型；当前保留回退接口。")
+              for capability in ("transcription", "speaker_matching")],
+        ])
+
+    def update(self, config: ModelRoutingConfig) -> ModelRoutingResponse:
+        for capability in CAPABILITIES:
+            binding = getattr(config, capability)
+            if binding:
+                try:
+                    provider = self.providers.get_any(binding.provider_id).config
+                except ProviderNotFoundError as exc:
+                    raise ApiError(422, "PROVIDER_NOT_FOUND", "请选择已保存的提供商。") from exc
+                if provider.provider_type not in HTTP_TYPES:
+                    raise ApiError(422, "MODEL_ROUTING_PROTOCOL_UNSUPPORTED", "该能力当前需要 OpenAI Compatible HTTP 接口。")
+        conn = self._connection()
+        try:
+            with transaction(conn):
+                row = conn.execute("SELECT config_json FROM model_routing WHERE id=1").fetchone()
+                current = ModelRoutingConfig.model_validate_json(row[0]) if row else ModelRoutingConfig()
+                if current.version != config.version:
+                    raise ApiError(409, "MODEL_ROUTING_VERSION_CONFLICT", "配置已更新，请重新加载后再保存。")
+                saved = config.model_copy(update={"version": config.version + 1})
+                conn.execute("INSERT OR REPLACE INTO model_routing VALUES (1, ?)", (saved.model_dump_json(),))
+        finally:
+            conn.close()
+        return self.describe()
+
+    def uses_provider(self, provider_id: str) -> bool:
+        config = self.configuration()
+        return any(binding and binding.provider_id == provider_id for binding in
+                   (getattr(config, name) for name in CAPABILITIES))
+
+    def _remote(self, binding: ModelBinding) -> tuple[str, dict[str, str]]:
+        try:
+            provider = self.providers.get(binding.provider_id).config
+        except ProviderNotFoundError as exc:
+            raise ProviderError("PROVIDER_UNAVAILABLE", "Configured provider is unavailable.") from exc
+        if provider.provider_type not in HTTP_TYPES:
+            raise ProviderError("PROVIDER_CAPABILITY_UNSUPPORTED", "Provider does not support this HTTP capability.")
+        try:
+            key = self.credentials.resolve(provider.credential_id)
+        except CredentialStoreError as exc:
+            raise ProviderError("PROVIDER_CREDENTIAL_UNAVAILABLE", "Provider credential is unavailable.") from exc
+        if provider.credential_id and not key:
+            raise ProviderError("PROVIDER_CREDENTIAL_MISSING", "Provider credential is not configured.")
+        url = (provider.base_url or "https://api.openai.com/v1").rstrip("/") + binding.endpoint
+        return url, {"Authorization": f"Bearer {key}"} if key else {}
+
+    async def _request(self, binding: ModelBinding, *, remote: tuple[str, dict[str, str]] | None = None, **kwargs) -> tuple[dict, str]:
+        url, headers = remote or self._remote(binding)
+        try:
+            async with httpx.AsyncClient(timeout=30, transport=self.transport) as client:
+                async with client.stream("POST", url, headers=headers, **kwargs) as response:
+                    response.raise_for_status()
+                    body = bytearray()
+                    async for chunk in response.aiter_bytes():
+                        body.extend(chunk)
+                        if len(body) > MAX_RESPONSE_BYTES:
+                            raise invalid_response()
+                    data = json.loads(body)
+        except httpx.TimeoutException as exc:
+            raise ProviderError("PROVIDER_TIMEOUT", "Model API timed out.") from exc
+        except httpx.HTTPStatusError as exc:
+            code = {401: "PROVIDER_AUTH_FAILED", 403: "PROVIDER_AUTH_FAILED", 404: "MODEL_NOT_FOUND", 429: "PROVIDER_RATE_LIMITED"}.get(exc.response.status_code, "PROVIDER_UNAVAILABLE")
+            raise ProviderError(code, f"Model API returned HTTP {exc.response.status_code}.") from exc
+        except (httpx.HTTPError, httpx.InvalidURL) as exc:
+            raise ProviderError("PROVIDER_UNAVAILABLE", "Model API is unavailable.") from exc
+        except (ValueError, UnicodeError) as exc:
+            raise invalid_response() from exc
+        if not isinstance(data, dict) or data.get("error"):
+            raise invalid_response()
+        return data, url
+
+    async def embed(self, texts: list[str]) -> EmbeddingResult:
+        config = self.configuration()
+        binding = config.embedding
+        record_embedding(route_version=config.version,
+                         requested_route=binding.model_dump() if binding else None)
+        reason = None
+        if binding and texts:
+            try:
+                vectors = []
+                dimension = binding.dimensions
+                # Freeze the origin across batches, even if the user edits the provider.
+                remote = self._remote(binding)
+                for start in range(0, len(texts), 32):
+                    batch = texts[start:start + 32]
+                    payload = {"model": binding.model, "input": batch, "encoding_format": "float"}
+                    if binding.dimensions is not None:
+                        payload["dimensions"] = binding.dimensions
+                    data, url = await self._request(binding, remote=remote, json=payload)
+                    items = data.get("data")
+                    if not isinstance(items, list) or len(items) != len(batch):
+                        raise invalid_response()
+                    indexed = {}
+                    for item in items:
+                        if not isinstance(item, dict):
+                            raise invalid_response()
+                        index, vector = item.get("index"), item.get("embedding")
+                        if type(index) is not int or index in indexed or not 0 <= index < len(batch):
+                            raise invalid_response()
+                        if not isinstance(vector, list) or not 1 <= len(vector) <= 16384:
+                            raise invalid_response()
+                        if any(not finite_number(value) for value in vector):
+                            raise invalid_response()
+                        dimension = dimension or len(vector)
+                        norm = math.hypot(*vector)
+                        if len(vector) != dimension or not norm or not math.isfinite(norm):
+                            raise invalid_response()
+                        indexed[index] = [value / norm for value in vector]
+                    vectors.extend(indexed[index] for index in range(len(batch)))
+                identity = json.dumps([url, binding.model, dimension], separators=(",", ":"))
+                return EmbeddingResult(vectors=vectors, source="api", dimensions=dimension,
+                                       model_id="api-" + hashlib.sha256(identity.encode()).hexdigest())
+            except ProviderError as exc:
+                reason = exc.code
+        vectors = await self.local_embedding.embed_documents(texts)
+        return EmbeddingResult(vectors=vectors, source="local", model_id=self.local_embedding.model_id,
+                               dimensions=self.local_embedding.dim, fallback_reason=reason)
+
+    @staticmethod
+    def _media_file(path: Path):
+        try:
+            handle = path.open("rb")
+        except OSError as exc:
+            raise ApiError(404, "ATTACHMENT_NOT_FOUND", "Audio attachment was not found.") from exc
+        import os
+        if not 0 < os.fstat(handle.fileno()).st_size <= MAX_MEDIA_BYTES:
+            handle.close()
+            raise ApiError(413, "ATTACHMENT_TOO_LARGE", "Audio attachment must be between 1 byte and 25 MiB.")
+        return handle
+
+    async def transcribe(self, source: Path, language: str | None) -> RoutedTranscript:
+        binding = self.configuration().transcription
+        if binding is None:
+            with self._media_file(source):
+                pass
+        reason = None
+        if binding:
+            try:
+                fields = {"model": binding.model}
+                if language:
+                    fields["language"] = language
+                with self._media_file(source) as handle:
+                    data, _ = await self._request(binding, data=fields,
+                        files={"file": (source.name, handle, "application/octet-stream")})
+                text = data.get("text")
+                if not isinstance(text, str) or not text.strip():
+                    raise invalid_response()
+                return RoutedTranscript(text=text, source="api")
+            except ProviderError as exc:
+                reason = exc.code
+        try:
+            text = await self.local_speech.transcribe(source, language)
+            if not isinstance(text, str) or not text.strip():
+                raise ProviderError("LOCAL_MODEL_INVALID_RESPONSE", "Local transcription was empty.")
+            return RoutedTranscript(text=text, source="local", fallback_reason=reason)
+        except ProviderError as exc:
+            raise ApiError(503, exc.code, exc.message, {"fallback_reason": reason}) from exc
+
+    async def match_speakers(self, source: Path, reference: Path) -> SpeakerMatchResult:
+        binding = self.configuration().speaker_matching
+        if binding is None:
+            with self._media_file(source), self._media_file(reference):
+                pass
+        reason = None
+        if binding:
+            try:
+                # Explicit application contract, not an OpenAI-standard endpoint.
+                with self._media_file(source) as audio, self._media_file(reference) as sample:
+                    data, _ = await self._request(binding, data={"model": binding.model}, files={
+                        "file": (source.name, audio, "application/octet-stream"),
+                        "reference_file": (reference.name, sample, "application/octet-stream"),
+                    })
+                score = data.get("score")
+                if not finite_number(score) or not 0 <= score <= 1:
+                    raise invalid_response()
+                return SpeakerMatchResult(score=score, source="api")
+            except ProviderError as exc:
+                reason = exc.code
+        try:
+            score = await self.local_speech.match(source, reference)
+            if not finite_number(score) or not 0 <= score <= 1:
+                raise ProviderError("LOCAL_MODEL_INVALID_RESPONSE", "Local speaker matching was invalid.")
+            return SpeakerMatchResult(score=score, source="local", fallback_reason=reason)
+        except ProviderError as exc:
+            raise ApiError(503, exc.code, exc.message, {"fallback_reason": reason}) from exc

backend/app/providers/tool_names.py

@@ -0,0 +1,47 @@
+"""Keep internal namespaced tools compatible with providers' 64-character names."""
+import hashlib
+import re
+from functools import wraps
+
+from app.contracts import MessageRole, ModelRequest
+
+
+def prepare_tool_names(request: ModelRequest) -> tuple[ModelRequest, dict[str, str]]:
+    names = {tool.name for tool in request.tools}
+    for message in request.messages:
+        names.update(call.name for call in message.tool_calls)
+        if message.role == MessageRole.tool and message.name:
+            names.add(message.name)
+    mapping = {name: name for name in names if re.fullmatch(r"[A-Za-z0-9_-]{1,64}", name)}
+    used = set(mapping)
+    for name in sorted(names - mapping.keys()):
+        salt = 0
+        while True:
+            alias = "tool_" + hashlib.sha256(f"{name}:{salt}".encode()).hexdigest()[:56]
+            if alias not in used:
+                break
+            salt += 1
+        mapping[name] = alias
+        used.add(alias)
+    if all(name == alias for name, alias in mapping.items()):
+        return request, {}
+    wire = request.model_copy(deep=True)
+    for tool in wire.tools:
+        tool.name = mapping[tool.name]
+    for message in wire.messages:
+        for call in message.tool_calls:
+            call.name = mapping[call.name]
+        if message.role == MessageRole.tool and message.name:
+            message.name = mapping[message.name]
+    return wire, {alias: name for name, alias in mapping.items()}
+
+
+def mapped_tool_names(complete):
+    @wraps(complete)
+    async def wrapped(self, request: ModelRequest):
+        wire, originals = prepare_tool_names(request)
+        turn = await complete(self, wire)
+        for call in turn.tool_calls:
+            call.name = originals.get(call.name, call.name)
+        return turn
+    return wrapped

backend/app/repository.py

@@ -273,11 +273,15 @@ def update_note_location(
         raise LookupError(note_id)
 
 
-def fts_search_page(
+_FTS_FROM = """
-    *,
+        FROM blocks_fts
+        JOIN blocks AS b ON b.block_id = blocks_fts.block_id
+        JOIN notes AS n ON n.note_id = b.note_id
+    """
+
+
+def _fts_where(
     match: str,
-    limit: int,
-    offset: int,
     folders: list[str],
     note_ids: list[str],
     tags: list[str],
@@ -285,8 +289,11 @@ def fts_search_page(
     created_to: datetime | None,
     updated_from: datetime | None,
     updated_to: datetime | None,
-) -> tuple[list[FtsHit], int]:
+) -> tuple[str, list[object]]:
-    """执行带元数据过滤的 FTS 精确分页，并返回过滤后的完整命中数。"""
+    """构建 FTS 过滤 WHERE 子句（不含 WHERE 关键字），返回 (where_sql, params)。
+
+    fts_search_page 与 fts_score_bounds 共用，保证计数与取数口径一致。
+    """
     where = ["blocks_fts MATCH ?"]
     params: list[object] = [match]
 
@@ -317,22 +324,44 @@ def fts_search_page(
             where.append(f"julianday({column}) <= julianday(?)")
             params.append(_iso(upper))
 
-    from_sql = """
+    return " AND ".join(where), params
-        FROM blocks_fts
+
-        JOIN blocks AS b ON b.block_id = blocks_fts.block_id
+
-        JOIN notes AS n ON n.note_id = b.note_id
+def fts_search_page(
+    *,
+    match: str,
+    limit: int,
+    offset: int,
+    folders: list[str],
+    note_ids: list[str],
+    tags: list[str],
+    created_from: datetime | None,
+    created_to: datetime | None,
+    updated_from: datetime | None,
+    updated_to: datetime | None,
+    bm25_max: float | None = None,
+) -> tuple[list[FtsHit], int]:
+    """执行带元数据过滤的 FTS 精确分页，并返回过滤后的完整命中数。
+
+    bm25_max 非空时按 bm25 截止值过滤（用于阈值过滤的精确分页），计数与取数同口径。
     """
-    where_sql = " AND ".join(where)
+    where_sql, params = _fts_where(
+        match, folders, note_ids, tags,
+        created_from, created_to, updated_from, updated_to,
+    )
+    if bm25_max is not None:
+        where_sql += " AND bm25(blocks_fts) <= ?"
+        params.append(bm25_max)
 
     conn = connect()
     try:
         total = conn.execute(
-            f"SELECT COUNT(*) {from_sql} WHERE {where_sql}", params
+            f"SELECT COUNT(*) {_FTS_FROM} WHERE {where_sql}", params
         ).fetchone()[0]
         rows = conn.execute(
             f"""
             SELECT blocks_fts.block_id, blocks_fts.note_id, bm25(blocks_fts) AS rank
-            {from_sql}
+            {_FTS_FROM}
             WHERE {where_sql}
             ORDER BY rank
             LIMIT ? OFFSET ?
@@ -348,6 +377,45 @@ def fts_search_page(
         conn.close()
 
 
+def fts_score_bounds(
+    *,
+    match: str,
+    folders: list[str],
+    note_ids: list[str],
+    tags: list[str],
+    created_from: datetime | None,
+    created_to: datetime | None,
+    updated_from: datetime | None,
+    updated_to: datetime | None,
+) -> tuple[float, float] | None:
+    """返回 metadata 过滤后的 FTS 命中集里 bm25 的 (min, max)，无命中时返回 None。
+
+    用于阈值过滤：min-max 归一化是 bm25 的线性函数，据此可把阈值换算为 bm25 截止值。
+    """
+    where_sql, params = _fts_where(
+        match, folders, note_ids, tags,
+        created_from, created_to, updated_from, updated_to,
+    )
+    conn = connect()
+    try:
+        # bm25() 不能作为聚合函数参数，也不能用在被聚合的子查询里；改用 ORDER BY 取首尾两行
+        lo_row = conn.execute(
+            f"SELECT bm25(blocks_fts) AS rank {_FTS_FROM} WHERE {where_sql}"
+            " ORDER BY rank ASC LIMIT 1",
+            params,
+        ).fetchone()
+        if lo_row is None or lo_row["rank"] is None:
+            return None
+        hi_row = conn.execute(
+            f"SELECT bm25(blocks_fts) AS rank {_FTS_FROM} WHERE {where_sql}"
+            " ORDER BY rank DESC LIMIT 1",
+            params,
+        ).fetchone()
+        return (float(lo_row["rank"]), float(hi_row["rank"]))
+    finally:
+        conn.close()
+
+
 def get_block_hits(block_ids: list[str]) -> list[BlockHit]:
     if not block_ids:
         return []
@@ -392,16 +460,18 @@ def get_index_meta() -> dict[str, str]:
         conn.close()
 
 
-def clear_all() -> None:
+def clear_all(*, conn: sqlite3.Connection | None = None) -> None:
-    """清空元数据、Block 与 FTS5（重建索引用，向量由 VectorStore.clear 处理）。"""
+    """Clear rebuildable metadata using the caller's transaction when provided."""
-    conn = connect()
+    owns = conn is None
+    conn = conn or connect()
     try:
-        with transaction(conn):
+        with transaction(conn) if owns else nullcontext():
             conn.execute("DELETE FROM blocks_fts")
             conn.execute("DELETE FROM blocks")
             conn.execute("DELETE FROM notes")
     finally:
-        conn.close()
+        if owns:
+            conn.close()
 
 
 def stats() -> dict[str, int]:

backend/app/retrieval/embedding.py

@@ -19,6 +19,7 @@ class EmbeddingProvider(Protocol):
     """统一 Embedding 接口（与文档一致）。"""
 
     model_id: str
+    version: str
     dim: int
 
     async def embed_documents(self, texts: list[str]) -> list[list[float]]: ...
@@ -33,6 +34,7 @@ class HashEmbeddingProvider:
     """
 
     model_id = "hash-v1"
+    version = "1"
     dim = EMBEDDING_DIM
 
     async def embed_documents(self, texts: list[str]) -> list[list[float]]:

backend/app/retrieval/engine.py

@@ -22,6 +22,8 @@ from app.repository import BlockHit
 from app.retrieval.embedding import EmbeddingProvider, HashEmbeddingProvider
 from app.retrieval.hybrid import normalize_scores, rrf_fuse
 from app.retrieval.reranker import LexicalReranker, RankedCandidate, RerankerProvider
+from app.retrieval import routed_vectors
+from app.retrieval.provenance import record_embedding
 from app.retrieval.vectorstore import SqliteVecStore, VectorStore
 from app.textutils import make_snippet, match_query
 
@@ -39,10 +41,15 @@ class RetrievalEngine:
         embedding: EmbeddingProvider,
         reranker: RerankerProvider,
         vector_store: VectorStore,
+        *,
+        route_embeddings: bool = False,
     ) -> None:
         self.embedding = embedding
         self.reranker = reranker
         self.vector_store = vector_store
+        # Only the production instance opts in. Replaced test dependencies must
+        # remain authoritative, including monkeypatches on the singleton.
+        self._routed_defaults = (embedding, vector_store) if route_embeddings else None
 
     async def search(self, request: SearchRequest) -> SearchResponse:
         if request.mode == SearchMode.fts:
@@ -56,7 +63,7 @@ class RetrievalEngine:
         # 候选池至少覆盖本次请求的 offset+limit，保证分页能取到目标页；设上限防内存失控
         window = min(request.offset + request.limit, MAX_CANDIDATE_POOL)
         pool_size = max(CANDIDATE_POOL, window)
-        # 带过滤时放大召回；FTS 则一次性取全量命中（≤FTS_FETCH_LIMIT）避免截断漏召回
+        # 带过滤时放大召回，缓解「先截断候选池再过滤」造成的漏召回
         recall = min(pool_size * OVERSCAN_FACTOR, MAX_CANDIDATE_POOL) if has_filters else pool_size
 
         # 1. 按模式收集候选（FTS 与 Vector 各产出「按相关性降序」的 block_id 列表）
@@ -74,8 +81,19 @@ class RetrievalEngine:
                 fts_scores = {h.block_id: -h.bm25 for h in fts_hits}
 
         if request.mode in (SearchMode.vector, SearchMode.hybrid):
-            query_vec = await self.embedding.embed_query(request.query)
+            record_embedding(source="unavailable")
-            vec_hits = await self.vector_store.search(query_vec, top_k=recall)
+            vec_hits = None
+            if (
+                self._routed_defaults is not None
+                and self.embedding is self._routed_defaults[0]
+                and self.vector_store is self._routed_defaults[1]
+            ):
+                vec_hits = await routed_vectors.search_remote(request.query, top_k=recall)
+            if vec_hits is None:
+                query_vec = await self.embedding.embed_query(request.query)
+                vec_hits = await self.vector_store.search(query_vec, top_k=recall)
+                record_embedding(source="local", model_id=self.embedding.model_id,
+                                 dimensions=self.embedding.dim, version=self.embedding.version)
             vec_ranked = [v.id for v in vec_hits]
             vec_scores = {v.id: v.score for v in vec_hits}
 
@@ -84,7 +102,7 @@ class RetrievalEngine:
         elif request.mode == SearchMode.vector:
             candidate_scores = vec_scores
         else:  # hybrid：RRF 融合
-            candidate_scores = rrf_fuse([fts_ranked, vec_ranked])
+            candidate_scores = rrf_fuse([fts_ranked, vec_ranked], k=request.rrf_k)
 
         if not candidate_scores:
             return self._empty(request)
@@ -97,14 +115,23 @@ class RetrievalEngine:
         if not filtered:
             return self._empty(request)
 
-        # 4. 排序 / 精排
+        # 4. 排序 / 精排：hybrid 先按融合分预排序，再对前 rerank_candidates 个候选做精排，
+        #    剩余候选按融合分排在精排结果之后；rerank=False 时跳过精排直接按融合分排序。
         if request.mode == SearchMode.hybrid:
-            candidates = [
+            pre_sorted = sorted(filtered, key=lambda h: -candidate_scores[h.block_id])
-                RankedCandidate(block_id=h.block_id, score=candidate_scores[h.block_id], text=h.content)
+            if request.rerank:
-                for h in filtered
+                limit = request.rerank_candidates
-            ]
+                pool = pre_sorted if limit is None else pre_sorted[:limit]
-            ranked = await self.reranker.rerank(request.query, candidates)
+                rest = [] if limit is None else pre_sorted[limit:]
-            ordered = [(c.block_id, c.score) for c in ranked]
+                candidates = [
+                    RankedCandidate(block_id=h.block_id, score=candidate_scores[h.block_id], text=h.content)
+                    for h in pool
+                ]
+                ranked = await self.reranker.rerank(request.query, candidates)
+                ordered = [(c.block_id, c.score) for c in ranked]
+                ordered += [(h.block_id, candidate_scores[h.block_id]) for h in rest]
+            else:
+                ordered = [(h.block_id, candidate_scores[h.block_id]) for h in pre_sorted]
         else:
             ordered = sorted(
                 ((h.block_id, candidate_scores[h.block_id]) for h in filtered),
@@ -112,8 +139,10 @@ class RetrievalEngine:
             )
 
         ordered = normalize_scores(ordered)
+        # score_threshold：归一化后过滤低分结果（默认 0 不过滤）
+        ordered = [(bid, score) for bid, score in ordered if score >= request.score_threshold]
 
-        # 5. 分页：total = 过滤后候选集大小。fts 已取全量（≤FTS_FETCH_LIMIT）故为真实命中数；
+        # 5. 分页：total = 过滤后候选集大小。fts 走数据库精确分页，total 为真实命中数；
         #    vector/hybrid 为 KNN 候选集，无全局 total。
         total = len(ordered)
         page = ordered[request.offset : request.offset + request.limit]
@@ -126,11 +155,41 @@ class RetrievalEngine:
         )
 
     def _search_fts(self, request: SearchRequest) -> SearchResponse:
-        """FTS 专用路径：过滤、COUNT 与分页全部在 SQLite 中完成。"""
+        """FTS 专用路径：在数据库侧完成过滤、计数与分页，不取全量后再截断。
+
+        阈值过滤时，min-max 归一化是 bm25 的线性函数，据此把 score_threshold 换算为
+        bm25 截止值（bm25_max），使过滤、计数与分页口径一致；无阈值时走数据库原生分页，
+        total 始终为过滤后的真实命中数，不再受固定截断影响。
+        """
         match = match_query(request.query)
         if not match:
             return self._empty(request)
 
+        bounds = repository.fts_score_bounds(
+            match=match,
+            folders=request.folders,
+            note_ids=request.note_ids,
+            tags=request.tags,
+            created_from=request.created_from,
+            created_to=request.created_to,
+            updated_from=request.updated_from,
+            updated_to=request.updated_to,
+        )
+        if bounds is None:
+            return self._empty(request)
+
+        lo, hi = bounds
+        span = hi - lo
+        bm25_max: float | None = None
+        if request.score_threshold > 0:
+            if span == 0:
+                # 全部命中 bm25 相同，归一化后皆为 1.0；阈值超过 1.0 时无命中
+                if request.score_threshold > 1.0:
+                    return self._empty(request)
+            else:
+                # norm = (hi - bm25) / span；norm >= threshold ⟺ bm25 <= hi - threshold * span
+                bm25_max = hi - request.score_threshold * span
+
         fts_hits, total = repository.fts_search_page(
             match=match,
             limit=request.limit,
@@ -142,19 +201,29 @@ class RetrievalEngine:
             created_to=request.created_to,
             updated_from=request.updated_from,
             updated_to=request.updated_to,
+            bm25_max=bm25_max,
         )
         if not fts_hits:
+            # 本页无结果：offset 越过末页时 total 仍为真实命中数（>0），需保留而非归零
             return SearchResponse(
                 query=request.query,
                 mode=request.mode,
+                items=[],
                 page=PageMeta(total=total, limit=request.limit, offset=request.offset),
             )
 
-        hits = {h.block_id: h for h in repository.get_block_hits([hit.block_id for hit in fts_hits])}
+        # 分数按全局 bm25 上下界归一化（与取全量后 normalize_scores 等价），保证跨页一致
-        ordered = normalize_scores(
+        span = hi - lo
-            [(hit.block_id, -hit.bm25) for hit in fts_hits if hit.block_id in hits]
+        if span == 0:
-        )
+            ordered = [(hit.block_id, 1.0) for hit in fts_hits]
-        items = [self._build_result(hits[block_id], request, score) for block_id, score in ordered]
+        else:
+            ordered = [(hit.block_id, round((hi - hit.bm25) / span, 6)) for hit in fts_hits]
+        hits = {h.block_id: h for h in repository.get_block_hits([bid for bid, _ in ordered])}
+        items = [
+            self._build_result(hits[block_id], request, score)
+            for block_id, score in ordered
+            if block_id in hits
+        ]
         return SearchResponse(
             query=request.query,
             mode=request.mode,
@@ -217,4 +286,6 @@ def _utc(dt: datetime) -> datetime:
 
 
 # 默认引擎实例：轻量实现跑通链路，后续可替换真实模型实现
-engine = RetrievalEngine(HashEmbeddingProvider(), LexicalReranker(), SqliteVecStore())
+engine = RetrievalEngine(
+    HashEmbeddingProvider(), LexicalReranker(), SqliteVecStore(), route_embeddings=True,
+)

backend/app/retrieval/provenance.py

@@ -0,0 +1,21 @@
+"""Task-local observations of the embedding path actually used by a search."""
+from contextlib import contextmanager
+from contextvars import ContextVar
+
+_observation: ContextVar[dict | None] = ContextVar("embedding_observation", default=None)
+
+
+@contextmanager
+def capture_embedding():
+    result = {"source": "not_used"}
+    token = _observation.set(result)
+    try:
+        yield result
+    finally:
+        _observation.reset(token)
+
+
+def record_embedding(**fields) -> None:
+    result = _observation.get()
+    if result is not None:
+        result.update(fields)

backend/app/retrieval/reranker.py

@@ -24,6 +24,7 @@ class RerankerProvider(Protocol):
     """统一 Reranker 接口：输入候选块，输出按相关性重排后的候选块。"""
 
     model_id: str
+    version: str
 
     async def rerank(self, query: str, candidates: list[RankedCandidate]) -> list[RankedCandidate]: ...
 
@@ -32,6 +33,7 @@ class LexicalReranker:
     """轻量精排：query 与块正文的词面重叠度，与归一化后的原始分数加权求和。"""
 
     model_id = "lexical-v1"
+    version = "1"
 
     def __init__(self, lexical_weight: float = 0.5) -> None:
         self.lexical_weight = lexical_weight

backend/app/retrieval/routed_vectors.py

@@ -0,0 +1,201 @@
+"""Optional API embeddings, isolated from the stable hash/sqlite-vec index.
+
+The runtime's model_id is the authoritative space ID (including provider URL,
+endpoint, model and dimensions); equal dimensions alone never imply compatibility.
+This phase uses a lazy, rebuildable SQLite side table instead of a schema migration.
+Search scans only current blocks in one database snapshot and requires complete
+coverage. Cosine ranking costs O(blocks * dimensions) with an O(top_k) heap; this
+small-vault implementation should become a per-space ANN index at larger scale.
+"""
+
+from __future__ import annotations
+
+import heapq
+import json
+import logging
+import math
+import sqlite3
+from dataclasses import dataclass
+from typing import Protocol
+
+from app.database.db import connect, transaction
+from app.retrieval.vectorstore import VectorHit
+from app.retrieval.provenance import record_embedding
+
+logger = logging.getLogger(__name__)
+
+
+class EmbeddingResult(Protocol):
+    vectors: list[list[float]]
+    source: str
+    model_id: str
+    dimensions: int
+    fallback_reason: str | None
+
+
+class EmbeddingRuntime(Protocol):
+    async def embed(self, texts: list[str]) -> EmbeddingResult: ...
+
+
+@dataclass(frozen=True)
+class RemoteEmbeddings:
+    space_id: str
+    dimensions: int
+    vectors: list[list[float]]
+
+
+def get_model_routing() -> EmbeddingRuntime | None:
+    """Lazy integration hook; tests can inject a runtime without any network I/O."""
+    from app.container import container
+
+    return getattr(container, "model_routing", None)
+
+
+def _unit_vector(vector: list[float], dimensions: int) -> list[float]:
+    if len(vector) != dimensions:
+        raise ValueError("embedding dimension mismatch")
+    if any(isinstance(value, bool) or not isinstance(value, (int, float)) for value in vector):
+        raise ValueError("embedding must be numeric")
+    if not all(math.isfinite(value) for value in vector):
+        raise ValueError("embedding must be finite")
+    scale = max(abs(value) for value in vector)
+    if scale == 0:
+        raise ValueError("embedding must be nonzero")
+    # Scaling first avoids overflow/underflow for finite but extreme API values.
+    scaled = [value / scale for value in vector]
+    norm = math.sqrt(math.fsum(value * value for value in scaled))
+    return [value / norm for value in scaled]
+
+
+async def embed_remote(texts: list[str]) -> RemoteEmbeddings | None:
+    """Return validated API vectors, or None to use the caller's local baseline.
+
+    Do not use the runtime's local result: the caller may have injected its own
+    embedding/store pair. Exception deliberately excludes cancellation.
+    """
+    if not texts:
+        return None
+    try:
+        runtime = get_model_routing()
+        if runtime is None:
+            return None
+        result = await runtime.embed(texts)
+        if result.source != "api":
+            record_embedding(fallback_reason=result.fallback_reason)
+            return None
+        if not isinstance(result.model_id, str) or not result.model_id or result.model_id == "hash-v1":
+            raise ValueError("API embedding needs a distinct space ID")
+        if type(result.dimensions) is not int or result.dimensions <= 0:
+            raise ValueError("invalid embedding dimensions")
+        if len(result.vectors) != len(texts):
+            raise ValueError("embedding count mismatch")
+        return RemoteEmbeddings(
+            space_id=result.model_id,
+            dimensions=result.dimensions,
+            vectors=[_unit_vector(vector, result.dimensions) for vector in result.vectors],
+        )
+    except Exception as exc:
+        # Avoid logging provider exceptions containing credentials or note text.
+        record_embedding(fallback_reason="REMOTE_EMBEDDING_UNAVAILABLE")
+        logger.warning("Remote embedding unavailable (%s); using local index", type(exc).__name__)
+        return None
+
+
+def _ensure_table(conn: sqlite3.Connection) -> None:
+    conn.execute("""
+        CREATE TABLE IF NOT EXISTS routed_block_vectors (
+            space_id TEXT NOT NULL,
+            block_id TEXT NOT NULL REFERENCES blocks(block_id) ON DELETE CASCADE,
+            dimensions INTEGER NOT NULL CHECK (dimensions > 0),
+            vector TEXT NOT NULL,
+            PRIMARY KEY (space_id, block_id)
+        )
+    """)
+    conn.execute("""
+        CREATE INDEX IF NOT EXISTS routed_block_vectors_block_id
+        ON routed_block_vectors(block_id)
+    """)
+
+
+def store_remote(
+    conn: sqlite3.Connection, block_ids: list[str], batch: RemoteEmbeddings | None,
+) -> None:
+    """Best-effort side-index write inside the caller's metadata transaction.
+
+    A savepoint prevents partial remote batches and isolates storage failures from
+    note saving. Replacing/deleting blocks cascades all old spaces automatically.
+    """
+    if batch is None:
+        return
+    try:
+        conn.execute("SAVEPOINT routed_vectors_write")
+        try:
+            if len(block_ids) != len(batch.vectors):
+                raise ValueError("block/vector count mismatch")
+            _ensure_table(conn)
+            conn.executemany(
+                """INSERT INTO routed_block_vectors (space_id, block_id, dimensions, vector)
+                   VALUES (?, ?, ?, ?)
+                   ON CONFLICT (space_id, block_id) DO UPDATE SET
+                       dimensions = excluded.dimensions, vector = excluded.vector""",
+                [
+                    (batch.space_id, block_id, batch.dimensions, json.dumps(vector, allow_nan=False))
+                    for block_id, vector in zip(block_ids, batch.vectors)
+                ],
+            )
+        except BaseException:
+            conn.execute("ROLLBACK TO routed_vectors_write")
+            raise
+        finally:
+            conn.execute("RELEASE routed_vectors_write")
+    except Exception as exc:
+        logger.warning("Remote vector storage unavailable (%s); local index retained", type(exc).__name__)
+
+
+async def search_remote(query: str, *, top_k: int) -> list[VectorHit] | None:
+    """None means fallback, including any missing/invalid current-block vector.
+
+    Read coverage and vectors together so concurrent note updates cannot produce
+    an apparently complete subset. Never fill missing remote hits with local hits.
+    """
+    batch = await embed_remote([query])
+    if batch is None:
+        return None
+    record_embedding(attempted_space={"model_id": batch.space_id, "dimensions": batch.dimensions})
+    try:
+        conn = connect()
+        try:
+            with transaction(conn):
+                exists = conn.execute(
+                    "SELECT 1 FROM sqlite_master WHERE type = 'table' AND name = 'routed_block_vectors'"
+                ).fetchone()
+                if exists is None:
+                    record_embedding(fallback_reason="REMOTE_INDEX_MISSING")
+                    return None
+                rows = conn.execute(
+                    """SELECT b.block_id, r.vector
+                       FROM blocks AS b
+                       LEFT JOIN routed_block_vectors AS r
+                         ON r.block_id = b.block_id AND r.space_id = ? AND r.dimensions = ?
+                       ORDER BY b.block_id""",
+                    (batch.space_id, batch.dimensions),
+                )
+
+                def hits():
+                    for row in rows:
+                        if row["vector"] is None:
+                            raise ValueError("remote space has incomplete block coverage")
+                        vector = _unit_vector(json.loads(row["vector"]), batch.dimensions)
+                        score = math.fsum(a * b for a, b in zip(batch.vectors[0], vector))
+                        yield VectorHit(id=row["block_id"], score=max(0.0, min(1.0, score)))
+
+                result = heapq.nlargest(top_k, hits(), key=lambda hit: hit.score)
+                record_embedding(source="api", model_id=batch.space_id,
+                                 dimensions=batch.dimensions, fallback_reason=None)
+                return result
+        finally:
+            conn.close()
+    except Exception as exc:
+        record_embedding(fallback_reason="REMOTE_INDEX_UNAVAILABLE")
+        logger.debug("Remote vector search unavailable (%s); using local index", type(exc).__name__)
+        return None

backend/app/retrieval/vectorstore.py

@@ -35,6 +35,7 @@ class VectorStore(Protocol):
     async def upsert(self, records: list[VectorRecord]) -> None: ...
     async def delete(self, ids: list[str]) -> None: ...
     async def search(self, vector: list[float], *, top_k: int) -> list[VectorHit]: ...
+    async def count(self) -> int: ...
 
 
 class SqliteVecStore:
@@ -85,10 +86,19 @@ class SqliteVecStore:
         finally:
             conn.close()
 
-    async def clear(self) -> None:
+    async def clear(self, *, conn: sqlite3.Connection | None = None) -> None:
-        conn = connect()
+        owns = conn is None
+        conn = conn or connect()
         try:
-            with transaction(conn):
+            with transaction(conn) if owns else nullcontext():
                 conn.execute("DELETE FROM vec_blocks")
+        finally:
+            if owns:
+                conn.close()
+
+    async def count(self) -> int:
+        conn = connect()
+        try:
+            return conn.execute("SELECT COUNT(*) FROM vec_blocks").fetchone()[0]
         finally:
             conn.close()

backend/app/routes.py

@@ -1,5 +1,6 @@
 import asyncio
 from collections.abc import AsyncIterator
+from contextlib import aclosing
 from datetime import datetime, timezone
 from uuid import uuid4
 
@@ -14,6 +15,14 @@ from app.contracts import (
     AgentRunListResponse,
     AgentTraceResponse,
     ChatRequest,
+    BenchmarkDatasetListResponse,
+    BenchmarkEventType,
+    BenchmarkKind,
+    BenchmarkReport,
+    BenchmarkRun,
+    BenchmarkRunListResponse,
+    BenchmarkStatus,
+    RAGRunRequest,
     CredentialStatus,
     CredentialWriteRequest,
     ExtensionInstallRequest,
@@ -33,6 +42,12 @@ from app.contracts import (
     McpToolSummaryListResponse,
     ModelEvent,
     ModelEventType,
+    EmbeddingRequest,
+    EmbeddingResult,
+    ModelRoutingConfig,
+    ModelRoutingResponse,
+    SpeakerMatchRequest,
+    SpeakerMatchResult,
     Note,
     NoteCreateRequest,
     NoteListResponse,
@@ -78,6 +93,10 @@ from app.contracts import (
     WorkspaceOpenRequest,
     WorkspaceSnapshot,
 )
+from app.agent import AgentCapacityError, AgentRunNotFoundError
+from app.benchmarks import datasets as benchmark_datasets
+from app.benchmarks import service as benchmark_service
+from app.container import container
 from app.errors import ApiError
 from app.extensions import ExtensionError
 from app.extensions.mcp_registry import McpRegistryError
@@ -96,10 +115,18 @@ from app.services import (
     transcription_service,
     workspace_service,
 )
+from app.services.attachment_service import attachment_path
 
 router = APIRouter(prefix="/api")
 
 
+@router.get("/permissions/policy", tags=["Permissions"])
+async def get_permission_policy() -> dict[str, str]:
+    from app.agent.permissions import KNOWN_PERMISSIONS
+    return {permission: container.permissions.policy.mode_for(permission).value
+            for permission in sorted(KNOWN_PERMISSIONS)}
+
+
 async def mcp_call_async(operation):
     """Even registry reads can wait on lifecycle locks; keep all MCP work off the event loop."""
     try:
@@ -294,17 +321,22 @@ async def chat(request: ChatRequest) -> StreamingResponse:
     provider = provider_or_404(request.provider_id)
 
     async def stream() -> AsyncIterator[str]:
+        sequence = 0
         try:
-            async for event in provider.adapter.stream(request):
+            async with aclosing(provider.adapter.stream(request)) as events:
-                yield as_sse(event.event.value, event.model_dump_json())
+                async for event in events:
-        except Exception as exc:
+                    sequence = event.sequence + 1
+                    yield as_sse(event.event.value, event.model_dump_json())
+        except Exception:
             error = ModelEvent(
                 event=ModelEventType.error,
-                data={"code": "PROVIDER_ERROR", "message": str(exc)},
+                sequence=sequence,
+                data={"code": "PROVIDER_ERROR", "message": "Provider could not complete the request."},
                 timestamp=utc_now(),
             )
             done = ModelEvent(
-                event=ModelEventType.done, sequence=1, timestamp=utc_now()
+                event=ModelEventType.done, sequence=sequence + 1,
+                data={"status": "failed"}, timestamp=utc_now()
             )
             yield as_sse(error.event.value, error.model_dump_json())
             yield as_sse(done.event.value, done.model_dump_json())
@@ -911,13 +943,13 @@ async def update_provider(
             409, "BUILTIN_PROVIDER_IMMUTABLE", "Mock provider cannot be modified."
         )
     fields = request.model_fields_set
-    if ("name" in fields and request.name is None) or (
+    if ("provider_type" in fields and request.provider_type is None) or ("name" in fields and request.name is None) or (
         "enabled" in fields and request.enabled is None
     ):
         raise ApiError(
             422,
             "VALIDATION_ERROR",
-            "name and enabled cannot be null when explicitly provided.",
+            "provider_type, name and enabled cannot be null when explicitly provided.",
         )
     updates = {name: getattr(request, name) for name in fields}
     if "credential_id" in fields:
@@ -925,7 +957,11 @@ async def update_provider(
     config = ProviderConfig.model_validate(
         {**current.model_dump(mode="python"), **updates}
     )
-    adapter = container.provider_factory.build(config)
+    config.capabilities = container.provider_factory.capabilities(config.provider_type)
+    try:
+        adapter = container.provider_factory.build(config)
+    except UnsupportedProviderError as exc:
+        raise ApiError(422, "PROVIDER_TYPE_UNSUPPORTED", "Provider adapter is not supported.") from exc
     container.providers.replace(config, adapter)
     return config
 
@@ -941,6 +977,8 @@ async def delete_provider(provider_id: str) -> OperationResponse:
         raise ApiError(
             409, "BUILTIN_PROVIDER_IMMUTABLE", "Mock provider cannot be deleted."
         )
+    if container.model_routing.uses_provider(provider_id):
+        raise ApiError(409, "PROVIDER_IN_USE", "请先在索引与模型中解除该提供商的模型绑定。")
     container.providers.unregister(provider_id)
     return OperationResponse(status="completed", resource_id=provider_id)
 
@@ -1043,6 +1081,28 @@ async def delete_task(task_id: str) -> OperationResponse:
 
 
 # Media and index
+@router.get("/model-routing", response_model=ModelRoutingResponse, tags=["Providers"])
+async def get_model_routing() -> ModelRoutingResponse:
+    return container.model_routing.describe()
+
+
+@router.put("/model-routing", response_model=ModelRoutingResponse, tags=["Providers"])
+async def update_model_routing(request: ModelRoutingConfig) -> ModelRoutingResponse:
+    return container.model_routing.update(request)
+
+
+@router.post("/models/embeddings", response_model=EmbeddingResult, tags=["Providers"])
+async def create_embeddings(request: EmbeddingRequest) -> EmbeddingResult:
+    return await container.model_routing.embed(request.texts)
+
+
+@router.post("/media/speaker-matches", response_model=SpeakerMatchResult, tags=["Media"])
+async def match_speakers(request: SpeakerMatchRequest) -> SpeakerMatchResult:
+    return await container.model_routing.match_speakers(
+        attachment_path(request.attachment_id), attachment_path(request.reference_attachment_id),
+    )
+
+
 @router.post(
     "/media/transcriptions",
     response_model=TranscriptionJob,
@@ -1050,8 +1110,8 @@ async def delete_task(task_id: str) -> OperationResponse:
     tags=["Media"],
 )
 async def create_transcription(request: TranscriptionRequest) -> TranscriptionJob:
-    return transcription_service.create_transcription(
+    return await transcription_service.create_transcription(
-        request.attachment_id, request.language
+        request.attachment_id, request.language, diarization=request.diarization
     )
 
 
@@ -1092,3 +1152,168 @@ async def get_index_job(job_id: str) -> IndexJob:
             404, "RESOURCE_NOT_FOUND", "index job not found", {"job_id": job_id}
         )
     return job
+
+
+# Benchmark
+@router.get(
+    "/benchmarks/datasets",
+    response_model=BenchmarkDatasetListResponse,
+    tags=["Benchmark"],
+)
+async def list_benchmark_datasets(
+    kind: BenchmarkKind = Query(default=BenchmarkKind.rag),
+) -> BenchmarkDatasetListResponse:
+    return BenchmarkDatasetListResponse(items=benchmark_datasets.list_datasets(kind))
+
+
+@router.post(
+    "/benchmarks/rag/runs",
+    response_model=BenchmarkRun,
+    status_code=202,
+    tags=["Benchmark"],
+)
+async def create_rag_benchmark(request: RAGRunRequest) -> BenchmarkRun:
+    return await benchmark_service.create_rag_run(request)
+
+
+@router.get(
+    "/benchmarks/runs",
+    response_model=BenchmarkRunListResponse,
+    tags=["Benchmark"],
+)
+async def list_benchmark_runs(
+    kind: BenchmarkKind | None = Query(default=None),
+    status: BenchmarkStatus | None = Query(default=None),
+    limit: int = Query(default=50, ge=1, le=100),
+    offset: int = Query(default=0, ge=0),
+) -> BenchmarkRunListResponse:
+    items, total = benchmark_service.list_runs(
+        kind=kind, status=status, limit=limit, offset=offset
+    )
+    return BenchmarkRunListResponse(
+        items=items, page=PageMeta(total=total, limit=limit, offset=offset)
+    )
+
+
+@router.get(
+    "/benchmarks/runs/{run_id}",
+    response_model=BenchmarkRun,
+    tags=["Benchmark"],
+)
+async def get_benchmark_run(run_id: str) -> BenchmarkRun:
+    run = benchmark_service.get_run(run_id)
+    if run is None:
+        raise ApiError(
+            404, "BENCHMARK_RUN_NOT_FOUND", "benchmark run not found", {"run_id": run_id}
+        )
+    return run
+
+
+@router.post(
+    "/benchmarks/runs/{run_id}/cancel",
+    response_model=OperationResponse,
+    tags=["Benchmark"],
+)
+async def cancel_benchmark_run(run_id: str) -> OperationResponse:
+    run = benchmark_service.cancel_run(run_id)
+    if run is None:
+        raise ApiError(
+            404, "BENCHMARK_RUN_NOT_FOUND", "benchmark run not found", {"run_id": run_id}
+        )
+    return OperationResponse(
+        status="accepted",
+        resource_id=run_id,
+        message=f"Benchmark run status: {run.status.value}",
+    )
+
+
+@router.get(
+    "/benchmarks/runs/{run_id}/events",
+    response_class=StreamingResponse,
+    responses={
+        200: {
+            "description": "BenchmarkEvent Server-Sent Events stream",
+            "content": {"text/event-stream": {}},
+        }
+    },
+    tags=["Benchmark"],
+)
+async def benchmark_events(
+    run_id: str,
+    after_sequence: int = Query(default=-1, ge=-1),
+    last_event_id: str | None = Header(default=None, alias="Last-Event-ID"),
+) -> StreamingResponse:
+    if benchmark_service.get_run(run_id) is None:
+        raise ApiError(
+            404, "BENCHMARK_RUN_NOT_FOUND", "benchmark run not found", {"run_id": run_id}
+        )
+
+    # SSE 断线重连：Last-Event-ID 优先于 after_sequence，用于从上次收到的事件继续
+    cursor = after_sequence
+    if last_event_id is not None:
+        try:
+            cursor = int(last_event_id)
+        except ValueError as exc:
+            raise ApiError(
+                400,
+                "BENCHMARK_EVENT_CURSOR_INVALID",
+                "Last-Event-ID must be an integer sequence.",
+                {"last_event_id": last_event_id},
+            ) from exc
+        if cursor < -1:
+            raise ApiError(
+                400,
+                "BENCHMARK_EVENT_CURSOR_INVALID",
+                "Last-Event-ID must be greater than or equal to -1.",
+            )
+
+    async def stream() -> AsyncIterator[str]:
+        # 先订阅（保证订阅之后产生的事件也能收到），再回放历史事件，最后实时输出新事件
+        terminal = (
+            BenchmarkEventType.run_completed,
+            BenchmarkEventType.run_failed,
+            BenchmarkEventType.run_cancelled,
+        )
+        queue = benchmark_service.subscribe(run_id)
+        try:
+            last_sequence = cursor
+            # 回放按订阅时刻的快照长度遍历，避免列表在回放期间被追加；终止事件同样要结束流，
+            # 防止回放完成后进入实时队列却因序号去重跳过同一终止事件而永久等待。
+            history = benchmark_service.get_events(run_id)
+            for index in range(len(history)):
+                event = history[index]
+                if event.sequence <= cursor:
+                    continue
+                yield as_sse(event.event.value, event.model_dump_json(), event_id=event.sequence)
+                last_sequence = event.sequence
+                if event.event in terminal:
+                    return
+            if queue is None:
+                return
+            while True:
+                event = await queue.get()
+                if event.sequence <= last_sequence:
+                    continue
+                yield as_sse(event.event.value, event.model_dump_json(), event_id=event.sequence)
+                last_sequence = event.sequence
+                if event.event in terminal:
+                    return
+        finally:
+            if queue is not None:
+                benchmark_service.unsubscribe(run_id, queue)
+
+    return StreamingResponse(stream(), media_type="text/event-stream")
+
+
+@router.get(
+    "/benchmarks/runs/{run_id}/report",
+    response_model=BenchmarkReport,
+    tags=["Benchmark"],
+)
+async def get_benchmark_report(run_id: str) -> BenchmarkReport:
+    report = benchmark_service.get_report(run_id)
+    if report is None:
+        raise ApiError(
+            404, "BENCHMARK_RUN_NOT_FOUND", "benchmark report not found", {"run_id": run_id}
+        )
+    return report

backend/app/services/index_service.py

@@ -6,7 +6,6 @@ MVP 阶段重建是同步的（数据量小），完成后直接返回 completed
 
 from __future__ import annotations
 
-import shutil
 from datetime import datetime, timezone
 from pathlib import Path
 from uuid import uuid4
@@ -16,8 +15,8 @@ from app.config import get_settings
 from app.contracts import IndexJob, IndexRebuildRequest, IndexStatus
 from app.errors import ApiError
 from app.knowledge.parser import parse_note
-from app.services.note_service import index_note
+from app.services.note_service import index_note, prepare_note_index
-from app.services import task_service
+from app.database.db import connect, transaction
 from app.services.coordination import serialized_vault_mutation
 from app.retrieval.vectorstore import SqliteVecStore
 
@@ -74,18 +73,7 @@ async def rebuild(request: IndexRebuildRequest) -> IndexJob:
             {"scope": request.scope, "note_ids": request.note_ids},
         )
 
-    # 先扫描到内存（失败不会清旧索引），再快照旧库用于失败回滚
     docs = _scan_vault()
-    settings = get_settings()
-    database_existed = settings.db_path.exists()
-    task_note_links = task_service.note_links() if database_existed else {}
-    backup_path = (
-        settings.db_path.with_name(f"{settings.db_path.name}.{job_id}.bak")
-        if database_existed
-        else None
-    )
-    if backup_path is not None:
-        shutil.copy2(settings.db_path, backup_path)
 
     _active_job_id = job_id
     _last_error = None
@@ -94,21 +82,34 @@ async def rebuild(request: IndexRebuildRequest) -> IndexJob:
         created_at=datetime.now(timezone.utc),
     ))
     try:
-        repository.clear_all()
+        prepared_notes = []
-        await vector_store.clear()
         for rel, folder, markdown, created, updated in docs:
             parsed = parse_note(
                 markdown=markdown, file_path=rel, folder=folder, tags=None,
                 created_at=created, updated_at=updated,
             )
-            await index_note(parsed)
+            prepared_notes.append((parsed, await prepare_note_index(parsed)))
-        task_service.restore_note_links(task_note_links)
+        # All network/model awaits precede the transaction. The concrete SQLite
+        # methods below complete synchronously despite their async interfaces.
+        conn = connect()
+        try:
+            with transaction(conn):
+                task_note_links = dict(conn.execute(
+                    "SELECT task_id, note_id FROM tasks WHERE note_id IS NOT NULL"
+                ).fetchall())
+                repository.clear_all(conn=conn)
+                await vector_store.clear(conn=conn)
+                for parsed, prepared in prepared_notes:
+                    await index_note(parsed, prepared=prepared, conn=conn)
+                for task_id, note_id in task_note_links.items():
+                    conn.execute(
+                        "UPDATE tasks SET note_id = ? WHERE task_id = ? "
+                        "AND EXISTS (SELECT 1 FROM notes WHERE note_id = ?)",
+                        (note_id, task_id, note_id),
+                    )
+        finally:
+            conn.close()
     except BaseException as exc:
-        # 重建失败：恢复旧索引，避免留下半成品；记录 failed 任务后向上抛
-        if backup_path is not None and backup_path.exists():
-            shutil.copy2(backup_path, settings.db_path)
-        elif not database_existed:
-            settings.db_path.unlink(missing_ok=True)
         _remember_job(IndexJob(
             job_id=job_id, status="failed", scope=request.scope,
             created_at=datetime.now(timezone.utc),
@@ -117,8 +118,6 @@ async def rebuild(request: IndexRebuildRequest) -> IndexJob:
         raise
     finally:
         _active_job_id = None
-        if backup_path is not None:
-            backup_path.unlink(missing_ok=True)
 
     job = IndexJob(job_id=job_id, status="completed", scope=request.scope, created_at=datetime.now(timezone.utc))
     _remember_job(job)
@@ -127,9 +126,12 @@ async def rebuild(request: IndexRebuildRequest) -> IndexJob:
 
 
 def get_status() -> IndexStatus:
+    counts = repository.stats()
     if _active_job_id is not None:
-        return IndexStatus(status="running", pending_jobs=0, active_job_id=_active_job_id)
+        return IndexStatus(status="running", pending_jobs=0, active_job_id=_active_job_id,
+                           total_notes=counts["notes"], total_blocks=counts["blocks"])
     return IndexStatus(
+        total_notes=counts["notes"], total_blocks=counts["blocks"],
         status="failed" if _last_error else "idle",
         pending_jobs=0,
         last_completed_at=_last_completed_at,

backend/app/services/note_service.py

@@ -6,6 +6,8 @@ Markdown 文件是笔记正文的持久化载体（Vault），SQLite/FTS5/向量
 
 from __future__ import annotations
 
+import sqlite3
+from contextlib import nullcontext
 from datetime import datetime, timezone
 from pathlib import Path
 from uuid import uuid4
@@ -16,6 +18,7 @@ from app.database.db import connect, transaction
 from app.errors import ApiError
 from app.knowledge.parser import ParsedNote, parse_note
 from app.retrieval.embedding import HashEmbeddingProvider
+from app.retrieval import routed_vectors
 from app.retrieval.vectorstore import SqliteVecStore, VectorRecord
 from app.services.coordination import serialized_vault_mutation
 from app.services.vault_paths import (
@@ -71,17 +74,34 @@ def _delete_markdown(rel_path: str) -> None:
         path.unlink()
 
 
-async def index_note(parsed: ParsedNote) -> None:
+PreparedIndex = tuple[list[list[float]], routed_vectors.RemoteEmbeddings | None]
+
+
+async def prepare_note_index(parsed: ParsedNote) -> PreparedIndex:
+    """Compute vectors before opening a write transaction (including API I/O)."""
+    texts = [block.content for block in parsed.blocks]
+    vectors = await embedding.embed_documents(texts)
+    remote = await routed_vectors.embed_remote(texts)
+    return vectors, remote
+
+
+async def index_note(
+    parsed: ParsedNote, *, prepared: PreparedIndex | None = None,
+    conn: sqlite3.Connection | None = None,
+) -> None:
     """把解析结果写入元数据 + FTS5 + 向量（三层可重建索引），单事务保证原子性。
 
     元数据与向量在同一连接、同一事务内提交，避免「新元数据已提交、向量写入失败」的
     半提交状态。替换元数据时拿到旧 block_id：清理已删除/内容变化的旧向量，只为新增
     block 写向量（内容未变的 block 其向量仍有效，无需重复写入）。
     """
-    vectors = await embedding.embed_documents([block.content for block in parsed.blocks])
+    if conn is not None and prepared is None:
-    conn = connect()
+        raise ValueError("Prepare embeddings before supplying a write connection")
+    vectors, remote = prepared if prepared is not None else await prepare_note_index(parsed)
+    owns = conn is None
+    conn = conn or connect()
     try:
-        with transaction(conn):
+        with transaction(conn) if owns else nullcontext():
             old_block_ids = repository.replace_note_metadata(
                 conn=conn,
                 note_id=parsed.note_id,
@@ -105,12 +125,14 @@ async def index_note(parsed: ParsedNote) -> None:
                 if block.block_id in missing_ids
             ]
             await vector_store.upsert(records, conn=conn)
+            routed_vectors.store_remote(conn, [block.block_id for block in parsed.blocks], remote)
             repository.set_index_meta(
                 {"embedding_model": embedding.model_id, "embedding_dim": str(embedding.dim)},
                 conn=conn,
             )
     finally:
-        conn.close()
+        if owns:
+            conn.close()
 
 
 @serialized_vault_mutation

backend/app/services/transcription_service.py

@@ -1,37 +1,59 @@
-"""转写适配层；第一阶段消费文本附件或桌面 Host 预生成的旁路文本。"""
+"""转写作业：API 优先，本地模型回退；保留已有 Host 文本入口。"""
 
 from __future__ import annotations
 
 from collections import OrderedDict
 from datetime import datetime, timezone
-from pathlib import Path
 from uuid import uuid4
 
 from app.contracts import TranscriptionJob
+from app.errors import ApiError
 from app.services.attachment_service import attachment_path
 
 _jobs: OrderedDict[str, TranscriptionJob] = OrderedDict()
 MAX_JOBS = 100
 
 
-def create_transcription(attachment_id: str, language: str | None = None) -> TranscriptionJob:
+async def create_transcription(attachment_id: str, language: str | None = None, *, diarization: bool = False) -> TranscriptionJob:
-    # TODO(ai-core): 第二阶段接入本地 ASR 队列后，保留相同 Job 契约替换此同步降级实现。
+    from app.container import container
-    del language  # 预生成 transcript 暂不需要语言识别。
+
     source = attachment_path(attachment_id)
-    transcript = source if source.suffix.lower() in {".txt", ".md"} else Path(f"{source}.txt")
     job = TranscriptionJob(
         job_id=f"transcription_{uuid4().hex}",
         attachment_id=attachment_id,
-        status="completed" if transcript.is_file() else "failed",
+        status="processing",
-        text=transcript.read_text(encoding="utf-8") if transcript.is_file() else None,
-        error_code=None if transcript.is_file() else "TRANSCRIPTION_BACKEND_UNAVAILABLE",
-        error_message=(
-            None
-            if transcript.is_file()
-            else "No host-generated transcript is available; local speech models are phase two."
-        ),
         created_at=datetime.now(timezone.utc),
     )
+    try:
+        if diarization:
+            # Speaker verification and diarization are different capabilities.
+            raise ApiError(501, "DIARIZATION_NOT_IMPLEMENTED", "说话人分离将在阶段 F 接入，当前不能忽略 diarization 请求。")
+        transcript = source if source.suffix.lower() in {".txt", ".md"} else attachment_path(f"{attachment_id}.txt")
+        # A saved transcript remains an explicit import path, never faked ASR.
+        if transcript.is_file() and (source == transcript or container.model_routing.configuration().transcription is None):
+            with transcript.open("rb") as handle:
+                content = handle.read(1024 * 1024 + 1)
+            if len(content) > 1024 * 1024:
+                raise ApiError(413, "TRANSCRIPT_TOO_LARGE", "Transcript exceeds 1 MiB.")
+            job.text = content.decode("utf-8")
+            if not job.text.strip():
+                raise ApiError(422, "TRANSCRIPT_EMPTY", "Transcript is empty.")
+            job.source = "sidecar"
+        else:
+            result = await container.model_routing.transcribe(source, language)
+            job.text = result.text
+            job.source = result.source
+            job.fallback_reason = result.fallback_reason
+        job.status = "completed"
+    except ApiError as exc:
+        job.status = "failed"
+        job.error_code = exc.code
+        job.error_message = exc.message
+        job.fallback_reason = exc.details.get("fallback_reason")
+    except (OSError, UnicodeError):
+        job.status = "failed"
+        job.error_code = "TRANSCRIPT_UNREADABLE"
+        job.error_message = "Transcript could not be read."
     _jobs[job.job_id] = job
     while len(_jobs) > MAX_JOBS:
         _jobs.popitem(last=False)

backend/data/benchmarks/rag-core-v1.json

@@ -0,0 +1,48 @@
+{
+  "dataset_id": "rag-core-v1",
+  "kind": "rag",
+  "version": "1.0.0",
+  "description": "基础中文笔记检索集（对应 backend/data/vault 内置语料，重建索引后即可复现）",
+  "cases": [
+    {
+      "case_id": "rag-vector-sim",
+      "query": "向量数据库如何进行相似度检索",
+      "expected_note_ids": ["note_c1454740a0e55ef5"],
+      "expected_block_ids": ["blk_07c4c6bce0ec4d12", "blk_605fb3593809f224"],
+      "citation_required": true,
+      "tags": ["向量数据库", "检索"]
+    },
+    {
+      "case_id": "rag-python-func",
+      "query": "Python 如何定义函数",
+      "expected_note_ids": ["note_424c3742c6f0e555"],
+      "expected_block_ids": ["blk_45d48cae2fed40fe", "blk_0768d9c25c2ecf07"],
+      "citation_required": true,
+      "tags": ["python"]
+    },
+    {
+      "case_id": "rag-citation",
+      "query": "搜索结果如何定位到原文位置",
+      "expected_note_ids": ["note_0c619caa30b1614c"],
+      "expected_block_ids": ["blk_3f6fcead71c25fc6", "blk_9af7b12e9ce909fc"],
+      "citation_required": true,
+      "tags": ["RAG"]
+    },
+    {
+      "case_id": "rag-hybrid",
+      "query": "混合检索怎么融合全文和向量",
+      "expected_note_ids": ["note_c1454740a0e55ef5"],
+      "expected_block_ids": ["blk_82b45418dba9f720"],
+      "citation_required": true,
+      "tags": ["检索"]
+    },
+    {
+      "case_id": "rag-tech-stack",
+      "query": "这个项目用什么后端和检索技术",
+      "expected_note_ids": ["note_3327e6cf18f3701f"],
+      "expected_block_ids": ["blk_feb2a9c42e7d31ad"],
+      "citation_required": false,
+      "tags": ["项目"]
+    }
+  ]
+}

backend/tests/test_benchmark.py

@@ -0,0 +1,567 @@
+"""Benchmark 服务的单元与端到端测试。
+
+沿用 conftest 的隔离机制：APP_DATA_DIR / DB / Vault 都指向临时目录，benchmark
+数据集也落在临时目录（settings.benchmark_datasets_path），不读写真实数据。
+
+运行采用「创建即 queued + 后台 Task 执行」的异步模型，测试通过 _run 在同一事件循环内
+创建并等待后台任务结束，得到终态 BenchmarkRun 后再断言。
+"""
+
+from __future__ import annotations
+
+import asyncio
+import json
+
+import pytest
+from pydantic import ValidationError
+
+from app.benchmarks import datasets, metrics as m, service
+from app.config import get_settings
+from app.contracts import (
+    BenchmarkKind,
+    BenchmarkRun,
+    BenchmarkStatus,
+    RAGRunRequest,
+    SearchMode,
+)
+from app.errors import ApiError
+
+
+def _write_dataset(dataset_id: str, cases: list[dict], *, kind: str = "rag") -> None:
+    directory = get_settings().benchmark_datasets_path
+    directory.mkdir(parents=True, exist_ok=True)
+    payload = {
+        "dataset_id": dataset_id,
+        "kind": kind,
+        "version": "1.0.0",
+        "description": "test dataset",
+        "cases": cases,
+    }
+    (directory / f"{dataset_id}.json").write_text(
+        json.dumps(payload, ensure_ascii=False), encoding="utf-8"
+    )
+
+
+def _write_raw(dataset_id: str, raw: dict) -> None:
+    directory = get_settings().benchmark_datasets_path
+    directory.mkdir(parents=True, exist_ok=True)
+    (directory / f"{dataset_id}.json").write_text(
+        json.dumps(raw, ensure_ascii=False), encoding="utf-8"
+    )
+
+
+def _run(request: RAGRunRequest):
+    """创建运行并在同一事件循环内等待后台任务结束，返回终态 BenchmarkRun。"""
+    from app.contracts import BenchmarkRun
+
+    async def _execute() -> BenchmarkRun:
+        run = await service.create_rag_run(request)
+        return await service.wait_for_run(run.run_id)
+
+    return asyncio.run(_execute())
+
+
+# --------------------------------------------------------------------------- #
+# 指标纯函数
+# --------------------------------------------------------------------------- #
+def test_hit_at_k_and_recall() -> None:
+    retrieved = ["a", "b", "c"]
+    expected = {"b", "z"}
+
+    assert m.hit_at_k(retrieved, expected, 1) is False
+    assert m.hit_at_k(retrieved, expected, 2) is True
+    assert m.recall_at_k(retrieved, expected, 5) == 0.5  # 只召回 b
+
+
+def test_recall_at_k_dedups_duplicate_notes() -> None:
+    # 同一 Note 经多个 Block 重复出现，去重后 Recall 不应超过 1
+    assert m.recall_at_k(["note-a", "note-a"], {"note-a"}, 2) == 1.0
+    assert m.recall_at_k(["note-a", "note-a", "note-b"], {"note-a"}, 3) == 1.0
+
+
+def test_reciprocal_rank_and_citation_hit() -> None:
+    assert m.reciprocal_rank(["x", "a", "b"], {"b"}) == 1 / 3
+    assert m.reciprocal_rank(["x"], {"b"}) == 0.0
+    assert m.citation_hit(["blk_1"], {"blk_1"}) is True
+    assert m.citation_hit(["blk_2"], {"blk_1"}) is False
+    assert m.citation_hit([], {"blk_1"}) is False
+
+
+def test_percentile() -> None:
+    assert m.percentile([1.0, 2.0, 3.0, 4.0], 50.0) == 2.5
+    assert m.percentile([], 50.0) == 0.0
+    assert m.percentile([7.0], 95.0) == 7.0
+
+
+# --------------------------------------------------------------------------- #
+# Dataset 注册与校验
+# --------------------------------------------------------------------------- #
+def test_list_datasets_empty_by_default() -> None:
+    assert datasets.list_datasets(BenchmarkKind.rag) == []
+
+
+def test_load_missing_dataset_raises() -> None:
+    with pytest.raises(ApiError) as exc:
+        datasets.load_dataset("does-not-exist", BenchmarkKind.rag)
+    assert exc.value.status_code == 404
+    assert exc.value.code == "BENCHMARK_DATASET_NOT_FOUND"
+
+
+def test_dataset_without_expected_ids_is_invalid() -> None:
+    _write_dataset("bad-v1", [{"case_id": "x", "query": "q", "citation_required": False}])
+    with pytest.raises(ApiError) as exc:
+        datasets.load_dataset("bad-v1", BenchmarkKind.rag)
+    assert exc.value.code == "BENCHMARK_DATASET_INVALID"
+
+
+def test_dataset_kind_mismatch_is_invalid() -> None:
+    _write_dataset("agent-v1", [{"case_id": "x", "query": "q", "expected_note_ids": ["n"]}], kind="agent")
+    with pytest.raises(ApiError) as exc:
+        datasets.load_dataset("agent-v1", BenchmarkKind.rag)
+    assert exc.value.code == "BENCHMARK_DATASET_INVALID"
+
+
+def test_citation_required_requires_expected_block_ids() -> None:
+    # citation_required=true 却没有 expected_block_ids，无法计算 Citation Hit Rate，应拒绝
+    _write_dataset(
+        "cit-req-v1",
+        [{"case_id": "x", "query": "q", "expected_note_ids": ["n"], "citation_required": True}],
+    )
+    with pytest.raises(ApiError) as exc:
+        datasets.load_dataset("cit-req-v1", BenchmarkKind.rag)
+    assert exc.value.code == "BENCHMARK_DATASET_INVALID"
+
+
+def test_list_datasets_skips_corrupted_structure() -> None:
+    # 合法 JSON 但字段结构错误（cases: 42），列表接口应隔离该文件而非整体 500
+    _write_raw("bad-structure", {"dataset_id": "bad-structure", "kind": "rag", "cases": 42})
+    _write_dataset("good-v1", [{"case_id": "x", "query": "q", "expected_note_ids": ["n"]}])
+
+    infos = datasets.list_datasets(BenchmarkKind.rag)
+    ids = {info.dataset_id for info in infos}
+    assert "good-v1" in ids
+    assert "bad-structure" not in ids
+
+
+# --------------------------------------------------------------------------- #
+# 请求校验（空 / 重复 modes）
+# --------------------------------------------------------------------------- #
+def test_empty_modes_rejected() -> None:
+    with pytest.raises(ValidationError):
+        RAGRunRequest(dataset_id="x", modes=[])
+
+
+def test_duplicate_modes_rejected() -> None:
+    with pytest.raises(ValidationError):
+        RAGRunRequest(dataset_id="x", modes=[SearchMode.fts, SearchMode.fts])
+
+
+# --------------------------------------------------------------------------- #
+# RAG Benchmark 端到端
+# --------------------------------------------------------------------------- #
+def _single_note_case() -> tuple[str, str, dict]:
+    from app.services import note_service
+
+    note = asyncio.run(
+        note_service.create_note(
+            title="向量库",
+            markdown="向量数据库用于存储高维向量并支持近似最近邻检索。",
+            folder="",
+            tags=["向量"],
+        )
+    )
+    case = {
+        "case_id": "c1",
+        "query": "向量数据库相似度检索",
+        "expected_note_ids": [note.note_id],
+        "expected_block_ids": [note.blocks[0].block_id],
+        "citation_required": True,
+        "tags": ["向量"],
+    }
+    return note.note_id, note.blocks[0].block_id, case
+
+
+def test_rag_benchmark_end_to_end() -> None:
+    _, _, case = _single_note_case()
+    _write_dataset("e2e-v1", [case])
+
+    run = _run(RAGRunRequest(dataset_id="e2e-v1", modes=[SearchMode.fts]))
+
+    assert run.status.value == "completed"
+    assert run.dataset_hash.startswith("sha256:")
+    assert run.metrics is not None
+
+    fts = run.metrics["fts"]
+    assert fts["hit_at_1"] == 1.0
+    assert fts["recall_at_k"] == 1.0
+    assert fts["mrr"] == 1.0
+    assert fts["citation_hit_rate"] == 1.0
+    assert fts["p50_latency_ms"] >= 0.0
+    assert fts["p95_latency_ms"] >= fts["p50_latency_ms"]
+
+
+def test_rag_benchmark_all_modes_produce_metrics() -> None:
+    _, _, case = _single_note_case()
+    _write_dataset("e2e-modes-v1", [case])
+
+    run = _run(RAGRunRequest(dataset_id="e2e-modes-v1"))
+    assert run.status.value == "completed"
+
+    for mode in ("fts", "vector", "hybrid"):
+        assert mode in run.metrics
+        for key in ("hit_at_1", "hit_at_5", "recall_at_k", "mrr", "citation_hit_rate"):
+            assert 0.0 <= run.metrics[mode][key] <= 1.0
+
+
+def test_config_snapshot_records_index_and_models() -> None:
+    _, _, case = _single_note_case()
+    _write_dataset("snapshot-v1", [case])
+
+    run = _run(RAGRunRequest(dataset_id="snapshot-v1", modes=[SearchMode.fts]))
+
+    snapshot = run.config_snapshot
+    assert snapshot["index_meta"] is not None
+    assert snapshot["embedding"]["policy"] == "per_case"
+    assert snapshot["local_embedding"]["version"]
+    assert snapshot["local_embedding"]["dim"]
+    assert snapshot["reranker"]["version"]
+    assert snapshot["retrieval"]["rrf_k"] == 60
+
+
+def test_benchmark_report_and_events() -> None:
+    _, _, case = _single_note_case()
+    _write_dataset("report-v1", [case])
+
+    run = _run(RAGRunRequest(dataset_id="report-v1", modes=[SearchMode.fts]))
+    report = service.get_report(run.run_id)
+    events = service.get_events(run.run_id)
+
+    assert report is not None
+    assert report.run_id == run.run_id
+    assert len(report.cases) == 1
+    assert report.cases[0].case_id == "c1"
+    assert report.cases[0].hit_at_1 is True
+
+    assert events, "运行应产生事件"
+    assert events[0].event.value == "RunStarted"
+    assert events[-1].event.value == "RunCompleted"
+
+
+def test_cancel_completed_run_keeps_status() -> None:
+    _, _, case = _single_note_case()
+    _write_dataset("cancel-v1", [case])
+
+    run = _run(RAGRunRequest(dataset_id="cancel-v1", modes=[SearchMode.fts]))
+    assert run.status.value == "completed"
+
+    cancelled = service.cancel_run(run.run_id)
+    assert cancelled.status.value == "completed"  # 已结束，不再变 cancelled
+
+
+def test_cancel_queued_run_marks_cancelled() -> None:
+    _, _, case = _single_note_case()
+    _write_dataset("cancel-queued-v1", [case])
+
+    async def _scenario():
+        run = await service.create_rag_run(
+            RAGRunRequest(dataset_id="cancel-queued-v1", modes=[SearchMode.fts])
+        )
+        service.cancel_run(run.run_id)
+        return await service.wait_for_run(run.run_id)
+
+    run = asyncio.run(_scenario())
+    assert run.status.value == "cancelled"
+
+
+# --------------------------------------------------------------------------- #
+# 指标聚合：Citation Hit Rate 只统计 citation_required 样本
+# --------------------------------------------------------------------------- #
+def test_citation_hit_rate_only_counts_citation_required() -> None:
+    from app.benchmarks import rag as rag_module
+    from app.contracts import RAGCaseResult
+
+    cases = [
+        RAGCaseResult(
+            case_id="a", mode=SearchMode.fts, repeat=0, latency_ms=1.0,
+            citation_hit=True, citation_applicable=True,
+        ),
+        RAGCaseResult(
+            case_id="b", mode=SearchMode.fts, repeat=0, latency_ms=1.0,
+            citation_hit=False, citation_applicable=False,
+        ),
+    ]
+    metrics = rag_module._aggregate(cases, SearchMode.fts)
+    # 只有 citation_applicable（citation_required=true）的样本计入分母
+    assert metrics.citation_hit_rate == 1.0
+
+
+# --------------------------------------------------------------------------- #
+# 路由接入
+# --------------------------------------------------------------------------- #
+def test_benchmark_routes_wired() -> None:
+    from app import routes
+
+    _, _, case = _single_note_case()
+    _write_dataset("route-v1", [case])
+
+    async def _scenario():
+        listed = await routes.list_benchmark_datasets(BenchmarkKind.rag)
+        assert any(item.dataset_id == "route-v1" for item in listed.items)
+
+        run = await routes.create_rag_benchmark(
+            RAGRunRequest(dataset_id="route-v1", modes=[SearchMode.fts])
+        )
+        assert run.status.value == "queued"
+        return await service.wait_for_run(run.run_id)
+
+    run = asyncio.run(_scenario())
+    assert run.status.value == "completed"
+
+    got = asyncio.run(routes.get_benchmark_run(run.run_id))
+    assert got.run_id == run.run_id
+
+    report = asyncio.run(routes.get_benchmark_report(run.run_id))
+    assert report.cases[0].case_id == "c1"
+
+
+def test_benchmark_run_not_found_raises() -> None:
+    from app import routes
+
+    with pytest.raises(ApiError) as exc:
+        asyncio.run(routes.get_benchmark_run("benchmark_missing"))
+    assert exc.value.code == "BENCHMARK_RUN_NOT_FOUND"
+
+
+# --------------------------------------------------------------------------- #
+# 审阅回归：索引兼容 / 容量 / 失败样本 / 取消事件 / 数据集隔离
+# --------------------------------------------------------------------------- #
+def test_create_rag_run_requires_built_index() -> None:
+    # 空索引（无已索引 block）会让所有模式得到全 0 指标，应在创建时拒绝而非跑出误导结果
+    _write_dataset("empty-index-v1", [{"case_id": "x", "query": "q", "expected_note_ids": ["n"]}])
+    with pytest.raises(ApiError) as exc:
+        asyncio.run(
+            service.create_rag_run(
+                RAGRunRequest(dataset_id="empty-index-v1", modes=[SearchMode.fts])
+            )
+        )
+    assert exc.value.status_code == 409
+    assert exc.value.code == "BENCHMARK_INDEX_INCOMPATIBLE"
+
+
+def test_capacity_exceeded_when_all_runs_active(monkeypatch) -> None:
+    # 满容量且全为活动（非终态）run 时，无法淘汰，应拒绝创建而非删掉正在运行的 run
+    _, _, case = _single_note_case()
+    _write_dataset("capacity-v1", [case])
+
+    monkeypatch.setattr(service, "MAX_RUNS", 1)
+    fake_id = "benchmark_fake_active"
+    service._runs[fake_id] = BenchmarkRun(
+        run_id=fake_id,
+        kind=BenchmarkKind.rag,
+        dataset_id="capacity-v1",
+        dataset_hash="sha256:fake",
+        status=BenchmarkStatus.queued,
+        created_at=service._now(),
+    )
+    try:
+        with pytest.raises(ApiError) as exc:
+            asyncio.run(
+                service.create_rag_run(
+                    RAGRunRequest(dataset_id="capacity-v1", modes=[SearchMode.fts])
+                )
+            )
+        assert exc.value.status_code == 429
+        assert exc.value.code == "BENCHMARK_CAPACITY_EXCEEDED"
+    finally:
+        service._runs.pop(fake_id, None)
+
+
+def test_failed_samples_counted_as_zero_in_aggregate() -> None:
+    from app.benchmarks import rag as rag_module
+    from app.contracts import RAGCaseResult
+
+    cases = [
+        RAGCaseResult(
+            case_id="ok", mode=SearchMode.fts, repeat=0, latency_ms=10.0,
+            hit_at_1=True, recall=1.0, reciprocal_rank=1.0,
+            citation_hit=True, citation_applicable=True,
+        ),
+        RAGCaseResult(
+            case_id="boom", mode=SearchMode.fts, repeat=0, latency_ms=0.0,
+            error="RAG case evaluation failed.",
+            error_code="BENCHMARK_CASE_EVALUATION_FAILED",
+        ),
+    ]
+    metrics = rag_module._aggregate(cases, SearchMode.fts)
+
+    assert metrics.total_cases == 2
+    assert metrics.successful_cases == 1
+    assert metrics.failed_cases == 1
+    assert metrics.failure_rate == 0.5
+    # 失败样本按零分计入质量指标分母，汇总不虚高
+    assert metrics.hit_at_1 == 0.5
+    assert metrics.recall_at_k == 0.5
+    # 延迟只统计成功样本
+    assert metrics.p50_latency_ms == 10.0
+
+
+def test_cancel_emits_run_cancelled_event() -> None:
+    _, _, case = _single_note_case()
+    _write_dataset("cancel-event-v1", [case])
+
+    async def _scenario():
+        run = await service.create_rag_run(
+            RAGRunRequest(dataset_id="cancel-event-v1", modes=[SearchMode.fts])
+        )
+        service.cancel_run(run.run_id)
+        return await service.wait_for_run(run.run_id)
+
+    run = asyncio.run(_scenario())
+    assert run.status.value == "cancelled"
+    events = service.get_events(run.run_id)
+    assert events[-1].event.value == "RunCancelled"
+
+
+def test_load_dataset_ignores_corrupted_unrelated_files() -> None:
+    # 无关文件损坏（非法 JSON / 顶层非对象）不应阻断目标数据集加载
+    directory = get_settings().benchmark_datasets_path
+    directory.mkdir(parents=True, exist_ok=True)
+    (directory / "broken.json").write_text("{ not valid json", encoding="utf-8")
+    (directory / "array.json").write_text('["a", "b"]', encoding="utf-8")
+    _write_dataset("ok-v1", [{"case_id": "x", "query": "q", "expected_note_ids": ["n"]}])
+
+    dataset = datasets.load_dataset("ok-v1", BenchmarkKind.rag)
+    assert dataset.dataset_id == "ok-v1"
+    assert len(dataset.cases) == 1
+
+
+def test_load_dataset_top_level_must_be_object() -> None:
+    _write_raw("array-top", ["a", "b"])
+    with pytest.raises(ApiError) as exc:
+        datasets.load_dataset("array-top", BenchmarkKind.rag)
+    assert exc.value.code == "BENCHMARK_DATASET_INVALID"
+
+
+# --------------------------------------------------------------------------- #
+# 审阅回归：运行中取消 / 仅块标注 / SSE 终止事件
+# --------------------------------------------------------------------------- #
+def test_cancel_running_benchmark_stops_early() -> None:
+    """运行中取消应在样本边界及时生效，而非跑完全部样本（审阅 P1）。"""
+    from app.benchmarks import service
+    from app.services import note_service
+
+    note = asyncio.run(
+        note_service.create_note(
+            title="取消回归", markdown="向量数据库用于存储高维向量。", folder="", tags=["向量"]
+        )
+    )
+    cases = [
+        {
+            "case_id": f"c{i}",
+            "query": "向量数据库",
+            "expected_note_ids": [note.note_id],
+            "expected_block_ids": [note.blocks[0].block_id],
+            "citation_required": True,
+        }
+        for i in range(50)
+    ]
+    _write_dataset("cancel-running-v1", cases)
+
+    async def _scenario():
+        run = await service.create_rag_run(
+            RAGRunRequest(dataset_id="cancel-running-v1", modes=[SearchMode.fts])
+        )
+
+        async def _cancel_after_start():
+            # 取消通过事件循环调度（独立 Task），而非同步直调，才能复现事件循环饥饿
+            while service.get_run(run.run_id).status == BenchmarkStatus.queued:
+                await asyncio.sleep(0)
+            service.cancel_run(run.run_id)
+
+        cancel_task = asyncio.create_task(_cancel_after_start())
+        finished = await service.wait_for_run(run.run_id)
+        await cancel_task
+        return finished
+
+    run = asyncio.run(_scenario())
+    assert run.status.value == "cancelled"
+    completed = sum(
+        1 for e in service.get_events(run.run_id) if e.event.value == "CaseCompleted"
+    )
+    assert completed < 50  # 未跑完全部样本，证明取消在样本边界生效
+
+
+def test_block_only_annotation_resolves_note_and_scores() -> None:
+    """仅标注 expected_block_ids 的样本应按块反查笔记评分，而非零分（审阅 P2）。"""
+    from app.services import note_service
+
+    note = asyncio.run(
+        note_service.create_note(
+            title="仅块标注", markdown="向量数据库存储高维向量。", folder="", tags=["向量"]
+        )
+    )
+    _write_dataset("block-only-v1", [{
+        "case_id": "c1",
+        "query": "向量数据库",
+        "expected_block_ids": [note.blocks[0].block_id],
+        "citation_required": False,
+    }])
+
+    run = _run(RAGRunRequest(dataset_id="block-only-v1", modes=[SearchMode.fts]))
+
+    assert run.status.value == "completed"
+    fts = run.metrics["fts"]
+    assert fts["hit_at_1"] == 1.0
+    assert fts["recall_at_k"] == 1.0
+    assert fts["mrr"] == 1.0
+
+
+def test_sse_stream_ends_on_terminal_event_in_replay() -> None:
+    """历史回放期间遇到终止事件时流应立即结束，而非进入实时队列永久等待（审阅 P2）。"""
+    from app import routes
+    from app.benchmarks import service
+    from app.contracts import BenchmarkEvent, BenchmarkEventType
+
+    run_id = "benchmark_sse_replay"
+    now = service._now()
+    # 模拟「回放期间运行完成」：run 仍为 running（subscribe 返回非空队列），
+    # 但历史事件里已含 RunCompleted 终止事件。
+    service._runs[run_id] = BenchmarkRun(
+        run_id=run_id,
+        kind=BenchmarkKind.rag,
+        dataset_id="d",
+        dataset_hash="sha256:x",
+        status=BenchmarkStatus.running,
+        created_at=now,
+    )
+    service._events[run_id] = [
+        BenchmarkEvent(
+            event=BenchmarkEventType.run_started, run_id=run_id, sequence=0,
+            data={}, timestamp=now,
+        ),
+        BenchmarkEvent(
+            event=BenchmarkEventType.run_completed, run_id=run_id, sequence=1,
+            data={}, timestamp=now,
+        ),
+    ]
+    try:
+        # 直调路由函数时 FastAPI 不解析 Query/Header 默认值，需显式传 None 覆盖 Header 哨兵
+        response = asyncio.run(
+            routes.benchmark_events(run_id, after_sequence=-1, last_event_id=None)
+        )
+
+        async def _collect() -> list[str]:
+            out: list[str] = []
+            async for chunk in response.body_iterator:
+                out.append(chunk)
+            return out
+
+        # 加超时防止回归（旧实现会永久挂起）
+        chunks = asyncio.run(asyncio.wait_for(_collect(), timeout=5))
+    finally:
+        service._forget(run_id)
+
+    events = [
+        line for chunk in chunks for line in chunk.splitlines() if line.startswith("event: ")
+    ]
+    assert events == ["event: RunStarted", "event: RunCompleted"]

backend/tests/test_live_frontend_metadata.py

@@ -0,0 +1,31 @@
+import asyncio
+
+from fastapi.testclient import TestClient
+
+from app.main import app
+from app.container import container
+from app.agent.permissions import PermissionMode
+from app.services.note_service import create_note
+
+
+def test_index_status_returns_real_counts():
+    with TestClient(app) as client:
+        initial = client.get('/api/index/status').json()
+        assert (initial['total_notes'], initial['total_blocks']) == (0, 0)
+        note = asyncio.run(create_note(title='Real note', markdown='# Real note\n\ncontent', folder=None, tags=[]))
+        result = client.get('/api/index/status').json()
+        assert result['total_notes'] == 1
+        assert result['total_blocks'] == len(note.blocks)
+
+
+def test_permissions_endpoint_reads_effective_backend_policy():
+    policy = container.permissions.policy
+    original = policy.mode_for('attachments.read')
+    try:
+        policy.set_rule('attachments.read', PermissionMode.deny)
+        with TestClient(app) as client:
+            response = client.get('/api/permissions/policy')
+        assert response.status_code == 200
+        assert response.json()['attachments.read'] == 'deny'
+    finally:
+        policy.set_rule('attachments.read', original)

backend/tests/test_model_routing.py

@@ -0,0 +1,663 @@
+"""Offline model-routing contracts, HTTP validation, media lifetimes and persistence.
+
+All HTTP uses MockTransport (or the in-process API). Credentials, models and
+attachments are fakes, and conftest redirects all storage to temporary paths.
+"""
+
+from __future__ import annotations
+
+import asyncio
+import hashlib
+import json
+from email import policy
+from email.parser import BytesParser
+from types import SimpleNamespace
+
+import httpx
+import pytest
+from fastapi.testclient import TestClient
+
+from app.contracts import ModelBinding, ModelRoutingConfig, ProviderConfig, ProviderType
+from app.errors import ApiError
+from app.providers import MockProvider
+from app.providers.credentials import CredentialStoreError
+from app.providers.registry import ProviderRegistry
+from app.providers.routing import ModelRoutingService, PendingSpeechBackend
+from app.retrieval.embedding import HashEmbeddingProvider
+
+
+def run(awaitable):
+    return asyncio.run(awaitable)
+
+
+def response(data, status=200):
+    # Raw JSON intentionally permits NaN/Infinity to exercise hostile API output.
+    return httpx.Response(status, content=json.dumps(data).encode(), headers={"content-type": "application/json"})
+
+
+class FakeCredentials:
+    def __init__(self):
+        self.value = "unit-test-placeholder"
+        self.error = None
+        self.calls = []
+
+    def resolve(self, credential_id):
+        self.calls.append(credential_id)
+        if self.error:
+            raise self.error
+        return self.value if credential_id else None
+
+
+class FakeEmbedding:
+    model_id = "fake-local-model"
+    dim = 3
+
+    def __init__(self):
+        self.calls = []
+        self.error = None
+
+    async def embed_documents(self, texts):
+        self.calls.append(list(texts))
+        if self.error:
+            raise self.error
+        return [[0.6, 0.8, 0.0] for _ in texts]
+
+
+class FakeSpeech:
+    available = True
+
+    def __init__(self):
+        self.calls = []
+        self.text = "local transcript"
+        self.score = 0.25
+        self.error = None
+
+    async def transcribe(self, source, language):
+        self.calls.append(("transcribe", source, language))
+        if self.error:
+            raise self.error
+        return self.text
+
+    async def match(self, source, reference):
+        self.calls.append(("match", source, reference))
+        if self.error:
+            raise self.error
+        return self.score
+
+
+@pytest.fixture(autouse=True)
+def no_real_http(monkeypatch):
+    async def reject_async(*args, **kwargs):
+        pytest.fail("Real HTTP transport is forbidden in model-routing tests")
+
+    def reject_sync(*args, **kwargs):
+        pytest.fail("Real HTTP transport is forbidden in model-routing tests")
+
+    monkeypatch.setattr(httpx.AsyncHTTPTransport, "handle_async_request", reject_async)
+    monkeypatch.setattr(httpx.HTTPTransport, "handle_request", reject_sync)
+
+
+@pytest.fixture
+def rig():
+    requests = []
+
+    def unexpected(request):
+        pytest.fail(f"Unexpected model HTTP request: {request.url}")
+
+    state = SimpleNamespace(handler=unexpected)
+
+    async def dispatch(request):
+        requests.append(request)
+        result = state.handler(request)
+        return await result if hasattr(result, "__await__") else result
+
+    providers = ProviderRegistry()
+    config = ProviderConfig(
+        provider_id="test-provider", provider_type=ProviderType.openai_compatible,
+        name="Fake provider", base_url="https://models.invalid/v1/", credential_id="test-credential",
+    )
+    providers.register(config, MockProvider())
+    credentials, embedding, speech = FakeCredentials(), FakeEmbedding(), FakeSpeech()
+    service = ModelRoutingService(
+        providers, credentials, local_embedding=embedding, local_speech=speech,
+        transport=httpx.MockTransport(dispatch),
+    )
+    return SimpleNamespace(
+        service=service, providers=providers, credentials=credentials,
+        embedding=embedding, speech=speech, requests=requests, http=state,
+    )
+
+
+def bind(rig, capability="embedding", **overrides):
+    endpoints = {
+        "embedding": "/embeddings", "transcription": "/audio/transcriptions",
+        "speaker_matching": "/audio/speaker-matches",
+    }
+    binding = ModelBinding(**{
+        "provider_id": "test-provider", "model": "test-model",
+        "endpoint": endpoints[capability], **overrides,
+    })
+    current = rig.service.configuration()
+    return rig.service.update(current.model_copy(update={capability: binding}))
+
+
+def assert_local(rig, result, texts, reason):
+    assert result.source == "local"
+    assert result.model_id == rig.embedding.model_id
+    assert result.dimensions == 3
+    assert result.vectors == [[0.6, 0.8, 0.0] for _ in texts]
+    assert result.fallback_reason == reason
+    assert rig.embedding.calls == [texts]
+
+
+def test_embedding_observation_keeps_request_binding_when_config_changes(rig):
+    from app.retrieval.provenance import capture_embedding
+    initial = bind(rig, model="original-model")
+
+    def handler(request):
+        assert json.loads(request.content)["model"] == "original-model"
+        bind(rig, model="next-model")
+        return response({"data": [{"index": 0, "embedding": [1, 0, 0]}]})
+
+    rig.http.handler = handler
+    with capture_embedding() as observation:
+        result = run(rig.service.embed(["query"]))
+    assert result.source == "api"
+    assert observation["route_version"] == initial.config.version
+    assert observation["requested_route"]["model"] == "original-model"
+    assert observation["requested_route"]["provider_id"] == "test-provider"
+    assert rig.service.configuration().embedding.model == "next-model"
+    assert rig.credentials.value not in json.dumps(observation)
+    assert "credential_id" not in json.dumps(observation)
+
+
+@pytest.fixture
+def audio(tmp_path):
+    source, reference = tmp_path / "audio.wav", tmp_path / "reference.wav"
+    source.write_bytes(b"fake-audio-content")
+    reference.write_bytes(b"fake-reference-content")
+    return source, reference
+
+
+def media_call(rig, capability, audio):
+    if capability == "transcription":
+        return rig.service.transcribe(audio[0], "zh")
+    return rig.service.match_speakers(*audio)
+
+
+def track_media_handles(rig, monkeypatch):
+    handles = []
+    original = rig.service._media_file
+
+    def tracked(path):
+        handle = original(path)
+        handles.append(handle)
+        return handle
+
+    monkeypatch.setattr(rig.service, "_media_file", tracked)
+    return handles
+
+
+def test_absent_binding_uses_hash_without_network(rig):
+    rig.service.local_embedding = HashEmbeddingProvider()
+    texts = ["hello retrieval", "向量检索"]
+    result = run(rig.service.embed(texts))
+    assert result.source == "local"
+    assert result.model_id == "hash-v1"
+    assert result.dimensions == 128
+    assert result.vectors == run(HashEmbeddingProvider().embed_documents(texts))
+    assert result.fallback_reason is None
+    assert rig.requests == rig.credentials.calls == []
+    statuses = {item.capability: item.status for item in rig.service.describe().local_backends}
+    assert statuses == {"embedding": "placeholder", "transcription": "ready", "speaker_matching": "ready"}
+
+
+def test_empty_embedding_input_does_not_call_remote(rig):
+    bind(rig)
+    result = run(rig.service.embed([]))
+    assert result.vectors == [] and result.source == "local"
+    assert rig.requests == []
+
+
+def test_remote_embedding_restores_batch_order_normalizes_and_sends_auth(rig):
+    bind(rig, dimensions=2)
+    texts = [str(index) for index in range(35)]
+
+    def handler(request):
+        assert request.method == "POST"
+        assert str(request.url) == "https://models.invalid/v1/embeddings"
+        assert request.headers["authorization"] == "Bearer unit-test-placeholder"
+        payload = json.loads(request.content)
+        assert payload["model"] == "test-model"
+        assert payload["dimensions"] == 2
+        assert payload["encoding_format"] == "float"
+        return response({"data": [
+            {"index": index, "embedding": [float(int(text) + 1), 1.0]}
+            for index, text in reversed(list(enumerate(payload["input"])))
+        ]})
+
+    rig.http.handler = handler
+    result = run(rig.service.embed(texts))
+    assert result.source == "api" and result.fallback_reason is None
+    assert result.dimensions == 2 and len(result.vectors) == 35
+    for index, vector in enumerate(result.vectors):
+        assert sum(value * value for value in vector) == pytest.approx(1.0)
+        assert vector[0] / vector[1] == pytest.approx(index + 1)
+    assert [json.loads(req.content)["input"] for req in rig.requests] == [texts[:32], texts[32:]]
+    assert rig.embedding.calls == []
+
+
+def test_space_id_is_stable_and_includes_full_url_model_and_inferred_dimensions(rig):
+    dimensions = 2
+
+    def handler(request):
+        assert "dimensions" not in json.loads(request.content)
+        return response({"data": [{"index": 0, "embedding": [1.0] * dimensions}]})
+
+    rig.http.handler = handler
+    bind(rig, model="  trimmed-model  ")
+
+    def check(url, model, dimension):
+        result = run(rig.service.embed(["hello"]))
+        digest = hashlib.sha256(json.dumps([url, model, dimension], separators=(",", ":")).encode()).hexdigest()
+        assert result.model_id == "api-" + digest
+        assert result.source == "api"
+        return result.model_id
+
+    first = check("https://models.invalid/v1/embeddings", "trimmed-model", 2)
+    assert first == check("https://models.invalid/v1/embeddings", "trimmed-model", 2)
+    config = rig.providers.get_any("test-provider").config.model_copy(update={"base_url": "https://models.invalid/v1"})
+    rig.providers.replace(config, MockProvider())
+    assert first == check("https://models.invalid/v1/embeddings", "trimmed-model", 2)
+    bind(rig, model="trimmed-model", endpoint="/custom/embeddings")
+    endpoint_id = check("https://models.invalid/v1/custom/embeddings", "trimmed-model", 2)
+    bind(rig, model="another-model", endpoint="/custom/embeddings")
+    model_id = check("https://models.invalid/v1/custom/embeddings", "another-model", 2)
+    dimensions = 3
+    dim_id = check("https://models.invalid/v1/custom/embeddings", "another-model", 3)
+    config = config.model_copy(update={"base_url": "https://other.invalid/v1"})
+    rig.providers.replace(config, MockProvider())
+    provider_id = check("https://other.invalid/v1/custom/embeddings", "another-model", 3)
+    assert len({first, endpoint_id, model_id, dim_id, provider_id}) == 5
+
+
+@pytest.mark.parametrize("data", [
+    {"data": []},
+    {"data": [{"index": 0, "embedding": [1, 0]}]},
+    {"data": [{"index": 0, "embedding": [1, 0]}, {"index": 0, "embedding": [0, 1]}]},
+    {"data": [{"index": 0, "embedding": [1, 0]}, {"index": 2, "embedding": [0, 1]}]},
+    {"data": [{"index": False, "embedding": [1, 0]}, {"index": 1, "embedding": [0, 1]}]},
+    {"data": [{"index": 0, "embedding": [1, 0]}, {"index": 1, "embedding": [0, 1, 0]}]},
+    {"data": [{"index": 0, "embedding": [1, 0]}, {"index": 1, "embedding": [float("nan"), 1]}]},
+    {"data": [{"index": 0, "embedding": [1, 0]}, {"index": 1, "embedding": [float("inf"), 1]}]},
+    {"data": [{"index": 0, "embedding": [1, 0]}, {"index": 1, "embedding": [True, 1]}]},
+    {"data": [{"index": 0, "embedding": [1, 0]}, {"index": 1, "embedding": [0, 0]}]},
+    {"data": [{"index": 0, "embedding": [1, 0]}, {"index": 1, "embedding": []}]},
+    {"data": [{"index": 0, "embedding": [1, 0]}, {"index": 1, "embedding": ["1", 0]}]},
+    {"data": [None, None]},
+    {"error": {"message": "in-band failure"}, "data": []},
+    [],
+], ids=["empty", "count", "duplicate-index", "out-of-range-index", "bool-index", "dimensions", "nan", "infinity", "bool", "zero", "empty-vector", "string", "invalid-items", "in-band-error", "non-object"])
+def test_invalid_remote_embeddings_fall_back_as_a_whole(rig, data):
+    bind(rig)
+    rig.http.handler = lambda request: response(data)
+    texts = ["first", "second"]
+    assert_local(rig, run(rig.service.embed(texts)), texts, "PROVIDER_INVALID_RESPONSE")
+
+
+def test_explicit_embedding_dimension_mismatch_falls_back(rig):
+    bind(rig, dimensions=3)
+    rig.http.handler = lambda request: response({"data": [{"index": 0, "embedding": [1, 0]}]})
+    assert_local(rig, run(rig.service.embed(["text"])), ["text"], "PROVIDER_INVALID_RESPONSE")
+
+
+def test_later_batch_dimension_mismatch_discards_earlier_remote_vectors(rig):
+    bind(rig)
+
+    def handler(request):
+        batch = json.loads(request.content)["input"]
+        dimension = 2 if len(rig.requests) == 1 else 3
+        return response({"data": [{"index": i, "embedding": [1] * dimension} for i in range(len(batch))]})
+
+    rig.http.handler = handler
+    texts = [str(i) for i in range(33)]
+    assert_local(rig, run(rig.service.embed(texts)), texts, "PROVIDER_INVALID_RESPONSE")
+    assert len(rig.requests) == 2
+
+
+@pytest.mark.parametrize("failure, reason", [
+    (401, "PROVIDER_AUTH_FAILED"), (403, "PROVIDER_AUTH_FAILED"),
+    (404, "MODEL_NOT_FOUND"), (429, "PROVIDER_RATE_LIMITED"), (500, "PROVIDER_UNAVAILABLE"),
+    ("timeout", "PROVIDER_TIMEOUT"), ("connect", "PROVIDER_UNAVAILABLE"),
+    ("json", "PROVIDER_INVALID_RESPONSE"),
+])
+def test_embedding_http_failures_use_injected_local(rig, failure, reason):
+    bind(rig)
+
+    def handler(request):
+        if failure == "timeout":
+            raise httpx.ReadTimeout("simulated timeout", request=request)
+        if failure == "connect":
+            raise httpx.ConnectError("simulated connection failure", request=request)
+        if failure == "json":
+            return httpx.Response(200, content=b"not JSON")
+        return response({"error": "failed"}, failure)
+
+    rig.http.handler = handler
+    assert_local(rig, run(rig.service.embed(["text"])), ["text"], reason)
+
+
+@pytest.mark.parametrize("failure, reason", [
+    ("missing-key", "PROVIDER_CREDENTIAL_MISSING"),
+    ("unreadable-key", "PROVIDER_CREDENTIAL_UNAVAILABLE"),
+    ("disabled-provider", "PROVIDER_UNAVAILABLE"),
+])
+def test_unavailable_remote_configuration_falls_back_without_http(rig, failure, reason):
+    bind(rig)
+    if failure == "missing-key":
+        rig.credentials.value = None
+    elif failure == "unreadable-key":
+        rig.credentials.error = CredentialStoreError("fake unavailable store")
+    else:
+        config = rig.providers.get_any("test-provider").config.model_copy(update={"enabled": False})
+        rig.providers.replace(config, MockProvider())
+    assert_local(rig, run(rig.service.embed(["text"])), ["text"], reason)
+    assert rig.requests == []
+
+
+@pytest.mark.parametrize("capability", ["transcription", "speaker_matching"])
+def test_media_success_sends_expected_multipart_and_closes_files(rig, audio, monkeypatch, capability):
+    bind(rig, capability)
+    handles = track_media_handles(rig, monkeypatch)
+
+    def handler(request):
+        assert request.headers["authorization"] == "Bearer unit-test-placeholder"
+        assert str(request.url).endswith("/audio/transcriptions" if capability == "transcription" else "/audio/speaker-matches")
+        message = BytesParser(policy=policy.default).parsebytes(
+            b"Content-Type: " + request.headers["content-type"].encode() + b"\r\nMIME-Version: 1.0\r\n\r\n" + request.content,
+        )
+        parts = {part.get_param("name", header="content-disposition"): part for part in message.iter_parts()}
+        assert parts["model"].get_payload(decode=True) == b"test-model"
+        assert parts["file"].get_filename() == audio[0].name
+        assert parts["file"].get_payload(decode=True) == audio[0].read_bytes()
+        if capability == "transcription":
+            assert set(parts) == {"model", "language", "file"}
+            assert parts["language"].get_payload(decode=True) == b"zh"
+            return response({"text": "remote transcript"})
+        assert set(parts) == {"model", "file", "reference_file"}
+        assert parts["reference_file"].get_filename() == audio[1].name
+        assert parts["reference_file"].get_payload(decode=True) == audio[1].read_bytes()
+        return response({"score": 0.875})
+
+    rig.http.handler = handler
+    result = run(media_call(rig, capability, audio))
+    assert result.source == "api" and result.fallback_reason is None
+    assert result.text == "remote transcript" if capability == "transcription" else result.score == 0.875
+    assert len(handles) == (1 if capability == "transcription" else 2)
+    assert all(handle.closed for handle in handles)
+    assert rig.speech.calls == []
+
+
+@pytest.mark.parametrize("capability, data", [
+    ("transcription", {}), ("transcription", {"text": "  "}), ("transcription", {"text": False}),
+    ("transcription", {"error": "in-band", "text": "must not use"}),
+    ("speaker_matching", {}), ("speaker_matching", {"score": -0.1}),
+    ("speaker_matching", {"score": 1.1}), ("speaker_matching", {"score": True}),
+    ("speaker_matching", {"score": float("nan")}), ("speaker_matching", {"score": "0.5"}),
+    ("speaker_matching", {"error": "in-band", "score": 0.9}),
+])
+def test_invalid_remote_media_falls_back_to_injected_local(rig, audio, monkeypatch, capability, data):
+    bind(rig, capability)
+    handles = track_media_handles(rig, monkeypatch)
+    rig.http.handler = lambda request: response(data)
+    result = run(media_call(rig, capability, audio))
+    assert result.source == "local" and result.fallback_reason == "PROVIDER_INVALID_RESPONSE"
+    assert result.text == "local transcript" if capability == "transcription" else result.score == 0.25
+    assert rig.speech.calls == [
+        ("transcribe", audio[0], "zh") if capability == "transcription" else ("match", *audio)
+    ]
+    assert handles and all(handle.closed for handle in handles)
+
+
+@pytest.mark.parametrize("capability", ["transcription", "speaker_matching"])
+@pytest.mark.parametrize("configured", [False, True])
+def test_pending_local_backend_has_explicit_503_and_fallback_details(rig, audio, capability, configured):
+    rig.service.local_speech = PendingSpeechBackend()
+    if configured:
+        bind(rig, capability)
+        rig.http.handler = lambda request: response({"error": "unauthorized"}, 401)
+    with pytest.raises(ApiError) as caught:
+        run(media_call(rig, capability, audio))
+    assert caught.value.status_code == 503
+    assert caught.value.code == "LOCAL_MODEL_NOT_INSTALLED"
+    assert caught.value.details == {"fallback_reason": "PROVIDER_AUTH_FAILED" if configured else None}
+    statuses = {item.capability: item.status for item in rig.service.describe().local_backends}
+    assert statuses["transcription"] == statuses["speaker_matching"] == "not_installed"
+    assert len(rig.requests) == int(configured)
+
+
+@pytest.mark.parametrize("capability", ["transcription", "speaker_matching"])
+def test_invalid_local_speech_returns_explicit_503(rig, audio, capability):
+    rig.speech.text = ""
+    rig.speech.score = True
+    with pytest.raises(ApiError) as caught:
+        run(media_call(rig, capability, audio))
+    assert (caught.value.status_code, caught.value.code) == (503, "LOCAL_MODEL_INVALID_RESPONSE")
+    assert caught.value.details == {"fallback_reason": None}
+
+
+@pytest.mark.parametrize("capability", ["embedding", "transcription", "speaker_matching"])
+@pytest.mark.parametrize("stage", ["remote", "local"])
+def test_cancellation_propagates_and_upload_handles_close(rig, audio, monkeypatch, capability, stage):
+    bind(rig, capability)
+    handles = track_media_handles(rig, monkeypatch)
+
+    async def cancelled(request):
+        raise asyncio.CancelledError()
+
+    if stage == "remote":
+        rig.http.handler = cancelled
+    else:
+        rig.http.handler = lambda request: response({"error": "fallback"}, 500)
+        rig.embedding.error = rig.speech.error = asyncio.CancelledError()
+    operation = rig.service.embed(["text"]) if capability == "embedding" else media_call(rig, capability, audio)
+    with pytest.raises(asyncio.CancelledError):
+        run(operation)
+    assert len(handles) == {"embedding": 0, "transcription": 1, "speaker_matching": 2}[capability]
+    assert all(handle.closed for handle in handles)
+    if stage == "remote":
+        assert rig.embedding.calls == rig.speech.calls == []
+
+
+def test_missing_reference_closes_already_open_source(rig, audio, monkeypatch):
+    bind(rig, "speaker_matching")
+    handles = track_media_handles(rig, monkeypatch)
+    audio[1].unlink()
+    with pytest.raises(ApiError) as caught:
+        run(rig.service.match_speakers(*audio))
+    assert caught.value.status_code == 404
+    assert len(handles) == 1 and handles[0].closed
+    assert rig.requests == []
+
+
+def test_config_optimistic_conflict_preserves_saved_bindings(rig):
+    assert rig.service.configuration().version == 0
+    saved = bind(rig).config
+    assert saved.version == 1
+    with pytest.raises(ApiError) as caught:
+        rig.service.update(ModelRoutingConfig(version=0))
+    assert (caught.value.status_code, caught.value.code) == (409, "MODEL_ROUTING_VERSION_CONFLICT")
+    assert rig.service.configuration() == saved
+    assert rig.service.uses_provider("test-provider")
+    assert not rig.service.uses_provider("not-a-provider")
+    cleared = rig.service.update(ModelRoutingConfig(version=1)).config
+    assert cleared.version == 2 and cleared.embedding is None
+    assert not rig.service.uses_provider("test-provider")
+
+
+@pytest.mark.parametrize("capability", ["embedding", "transcription", "speaker_matching"])
+@pytest.mark.parametrize("provider_id, code", [
+    ("missing", "PROVIDER_NOT_FOUND"), ("unsupported", "MODEL_ROUTING_PROTOCOL_UNSUPPORTED"),
+])
+def test_config_references_require_existing_supported_providers(rig, capability, provider_id, code):
+    rig.providers.register(
+        ProviderConfig(provider_id="unsupported", provider_type=ProviderType.ollama, name="unsupported"), MockProvider(),
+    )
+    with pytest.raises(ApiError) as caught:
+        bind(rig, capability, provider_id=provider_id)
+    assert (caught.value.status_code, caught.value.code) == (422, code)
+    assert rig.service.configuration() == ModelRoutingConfig()
+    assert rig.requests == []
+
+
+@pytest.fixture
+def api(monkeypatch, no_real_http, _isolate_data_dir):
+    # Import the production container only after temporary storage is configured.
+    from app import container as container_module, routes
+    from app.main import app
+
+    containers = []
+
+    def restart():
+        container = container_module.build_container()
+        container.model_routing.credentials = FakeCredentials()
+
+        def unexpected(request):
+            pytest.fail(f"Unexpected API-side provider HTTP: {request.url}")
+
+        container.model_routing.transport = httpx.MockTransport(unexpected)
+        monkeypatch.setattr(container_module, "container", container)
+        monkeypatch.setattr(routes, "container", container)
+        containers.append(container)
+        return container
+
+    container = restart()
+    client = TestClient(app)
+    yield SimpleNamespace(client=client, container=container, restart=restart)
+    client.close()
+    for container in containers:
+        container.plugins.shutdown()
+        container.mcp_servers.shutdown()
+
+
+def create_api_provider(api):
+    result = api.client.post("/api/providers", json={
+        "provider_type": "openai_compatible", "name": "Persisted fake",
+        "base_url": "https://persist.invalid/v1", "default_model": "fake-model",
+    })
+    assert result.status_code == 200, result.text
+    return result.json()
+
+
+def test_api_config_conflict_reference_delete_and_restart_persistence(api):
+    provider = create_api_provider(api)
+    provider_id = provider["provider_id"]
+    assert api.client.get("/api/model-routing").json()["config"]["version"] == 0
+    config = {"version": 0, "embedding": {"provider_id": provider_id, "model": "embed-model", "endpoint": "/embeddings"}}
+    saved = api.client.put("/api/model-routing", json=config)
+    assert saved.status_code == 200
+    assert saved.json()["config"]["version"] == 1
+    conflict = api.client.put("/api/model-routing", json=config)
+    assert conflict.status_code == 409
+    assert conflict.json()["error"]["code"] == "MODEL_ROUTING_VERSION_CONFLICT"
+    blocked = api.client.delete(f"/api/providers/{provider_id}")
+    assert blocked.status_code == 409 and blocked.json()["error"]["code"] == "PROVIDER_IN_USE"
+    restarted = api.restart()
+    assert restarted.providers.get_any(provider_id).config.model_dump(mode="json") == provider
+    assert api.client.get("/api/model-routing").json()["config"] == saved.json()["config"]
+    assert {item["provider_id"] for item in api.client.get("/api/providers").json()["items"]} == {"mock", provider_id}
+    cleared = api.client.put("/api/model-routing", json={"version": 1})
+    assert cleared.status_code == 200
+    assert api.client.delete(f"/api/providers/{provider_id}").status_code == 200
+    api.restart()
+    assert api.client.get(f"/api/providers/{provider_id}").status_code == 404
+    assert api.client.get("/api/model-routing").json()["config"]["version"] == 2
+
+
+def test_api_provider_type_patch_rebuilds_adapter_and_persists(api):
+    from app.providers.anthropic_messages import AnthropicMessagesProvider
+
+    provider = create_api_provider(api)
+    provider_id = provider["provider_id"]
+    changed = api.client.patch(f"/api/providers/{provider_id}", json={
+        "provider_type": "anthropic_messages", "base_url": "https://anthropic.invalid/v1",
+    })
+    assert changed.status_code == 200, changed.text
+    assert changed.json()["provider_type"] == "anthropic_messages"
+    assert changed.json()["name"] == provider["name"]
+    assert isinstance(api.container.providers.get_any(provider_id).adapter, AnthropicMessagesProvider)
+    restarted = api.restart()
+    assert isinstance(restarted.providers.get_any(provider_id).adapter, AnthropicMessagesProvider)
+    assert api.client.get(f"/api/providers/{provider_id}").json() == changed.json()
+    for invalid_type in (None, "mock", "nonexistent-type"):
+        rejected = api.client.patch(f"/api/providers/{provider_id}", json={"provider_type": invalid_type})
+        assert rejected.status_code == 422
+        assert api.client.get(f"/api/providers/{provider_id}").json() == changed.json()
+
+
+@pytest.mark.parametrize("endpoint", ["https://elsewhere.invalid/embed", "//elsewhere.invalid/embed", "relative", "/../embed", "/embed?key=test"])
+def test_api_config_rejects_non_provider_endpoint_paths(api, endpoint):
+    provider = create_api_provider(api)
+    result = api.client.put("/api/model-routing", json={
+        "embedding": {"provider_id": provider["provider_id"], "model": "embed", "endpoint": endpoint},
+    })
+    assert result.status_code == 422
+    assert api.client.get("/api/model-routing").json()["config"]["version"] == 0
+
+
+def test_api_embedding_reports_remote_and_fallback_sources(api):
+    provider = create_api_provider(api)
+    assert api.client.put("/api/model-routing", json={
+        "embedding": {"provider_id": provider["provider_id"], "model": "embed", "endpoint": "/embeddings"},
+    }).status_code == 200
+    api.container.model_routing.transport = httpx.MockTransport(
+        lambda request: response({"data": [{"index": 0, "embedding": [3, 4]}]}),
+    )
+    result = api.client.post("/api/models/embeddings", json={"texts": ["hello"]})
+    assert result.status_code == 200
+    assert result.json()["source"] == "api" and result.json()["vectors"][0] == pytest.approx([0.6, 0.8])
+    api.container.model_routing.transport = httpx.MockTransport(lambda request: response({"error": "denied"}, 401))
+    result = api.client.post("/api/models/embeddings", json={"texts": ["hello"]})
+    assert result.status_code == 200
+    assert result.json()["source"] == "local" and result.json()["model_id"] == "hash-v1"
+    assert result.json()["fallback_reason"] == "PROVIDER_AUTH_FAILED"
+    assert api.client.post("/api/models/embeddings", json={"texts": []}).status_code == 422
+
+
+def test_api_speech_failure_reports_reason_in_503_and_transcription_job(api):
+    from app.services.attachment_service import attachment_path
+
+    source, reference = attachment_path("audio.wav"), attachment_path("reference.wav")
+    source.parent.mkdir(parents=True, exist_ok=True)
+    source.write_bytes(b"test audio")
+    reference.write_bytes(b"test reference")
+    provider = create_api_provider(api)
+    assert api.client.put("/api/model-routing", json={
+        "transcription": {"provider_id": provider["provider_id"], "model": "asr", "endpoint": "/audio/transcriptions"},
+        "speaker_matching": {"provider_id": provider["provider_id"], "model": "voice", "endpoint": "/audio/speaker-matches"},
+    }).status_code == 200
+    api.container.model_routing.transport = httpx.MockTransport(lambda request: response({"error": "offline"}, 500))
+    match = api.client.post("/api/media/speaker-matches", json={"attachment_id": source.name, "reference_attachment_id": reference.name})
+    assert match.status_code == 503
+    assert match.json()["error"]["code"] == "LOCAL_MODEL_NOT_INSTALLED"
+    assert match.json()["error"]["details"] == {"fallback_reason": "PROVIDER_UNAVAILABLE"}
+    transcript = api.client.post("/api/media/transcriptions", json={"attachment_id": source.name, "language": "zh"})
+    assert transcript.status_code == 202
+    job = transcript.json()
+    assert job["status"] == "failed" and job["error_code"] == "LOCAL_MODEL_NOT_INSTALLED"
+    assert job["fallback_reason"] == "PROVIDER_UNAVAILABLE"
+    assert api.client.get(f"/api/media/transcriptions/{job['job_id']}").json() == job
+
+
+@pytest.mark.parametrize("capability", ["embedding", "speaker_matching"])
+def test_out_of_float_range_json_number_is_invalid_remote_and_falls_back(rig, audio, capability):
+    """JSON integers may be finite but too large to convert to a Python float."""
+    bind(rig, capability)
+    data = {"data": [{"index": 0, "embedding": [10 ** 400, 1]}]} if capability == "embedding" else {"score": 10 ** 400}
+    rig.http.handler = lambda request: response(data)
+    if capability == "embedding":
+        assert_local(rig, run(rig.service.embed(["text"])), ["text"], "PROVIDER_INVALID_RESPONSE")
+    else:
+        result = run(media_call(rig, capability, audio))
+        assert result.source == "local" and result.score == rig.speech.score
+        assert result.fallback_reason == "PROVIDER_INVALID_RESPONSE"

backend/tests/test_provider_adapters.py

@@ -84,7 +84,7 @@ def test_openai_compatible_maps_tool_call_and_credentials() -> None:
         )
     )
 
-    assert captured["tools"][0]["function"]["name"] == "math.add"
+    assert captured["tools"][0]["function"]["name"].startswith("tool_")
     assert turn.tool_calls[0].name == "math.add"
     assert turn.tool_calls[0].arguments == {"left": 1, "right": 2}
     assert turn.input_tokens == 8

backend/tests/test_provider_protocols.py

@@ -0,0 +1,610 @@
+"""Wire-level provider tests: no credentials, SDKs, clocks, or network services."""
+
+import asyncio
+import json
+
+import httpx
+import pytest
+
+from app.contracts import Message, MessageRole, ModelCapability, ModelEventType as E, ModelRequest, ToolCall, ToolDefinition
+from app.providers.anthropic_messages import AnthropicMessagesProvider
+from app.providers.base import ProviderError
+from app.providers.ollama import OllamaProvider
+from app.providers.openai_compatible import OpenAICompatibleProvider
+from app.providers.openai_responses import OpenAIResponsesProvider
+
+
+NATIVE = ["responses", "anthropic"]
+PROTOCOLS = [*NATIVE, "compatible", "ollama"]
+SECRET = "test-only-sensitive-upstream-body"
+
+
+class Credentials:
+    def resolve(self, credential_id):
+        return SECRET if credential_id else None
+
+
+class Bytes(httpx.AsyncByteStream):
+    def __init__(self, body: bytes, *, fragment: int = 17):
+        self.body = body
+        self.fragment = fragment
+        self.closed = False
+
+    async def __aiter__(self):
+        for offset in range(0, len(self.body), self.fragment):
+            yield self.body[offset:offset + self.fragment]
+
+    async def aclose(self):
+        self.closed = True
+
+
+class GatedBytes(Bytes):
+    def __init__(self, body):
+        super().__init__(body)
+        self.waiting = asyncio.Event()
+        self.release = asyncio.Event()
+
+    async def __aiter__(self):
+        yield self.body
+        self.waiting.set()
+        await self.release.wait()
+
+
+def provider(protocol, handler, *, credential_id="test"):
+    transport = httpx.MockTransport(handler)
+    if protocol == "ollama":
+        return OllamaProvider("https://provider.test", transport=transport)
+    cls = {"responses": OpenAIResponsesProvider, "anthropic": AnthropicMessagesProvider,
+           "compatible": OpenAICompatibleProvider}[protocol]
+    return cls("https://provider.test/v1/", credential_id, Credentials(), transport=transport)
+
+
+def request(*, history=False):
+    messages = [Message(role=MessageRole.user, content="查笔记")]
+    if history:
+        messages += [
+            Message(role=MessageRole.system, content="Additional rules"),
+            Message(role=MessageRole.assistant, content="Checking", tool_calls=[
+                ToolCall(tool_call_id="old_1", name="lookup", arguments={"query": "a"}),
+                ToolCall(tool_call_id="old_2", name="lookup", arguments={"query": "b"}),
+            ]),
+            Message(role=MessageRole.tool, tool_call_id="old_1", content='{"found":1}'),
+            Message(role=MessageRole.tool, tool_call_id="old_2", content='{"found":2}'),
+        ]
+    return ModelRequest(
+        provider_id="test", model="model", system="System rules", messages=messages,
+        tools=[ToolDefinition(name="lookup", description="Find notes", parameters={"type": "object"})],
+        max_tokens=512, temperature=0,
+    )
+
+
+async def collect(iterator):
+    return [event async for event in iterator]
+
+
+@pytest.mark.parametrize("name", ["lookup", "notes.search"])
+def test_compatible_split_tool_name_preserves_identity(name):
+    from app.providers.tool_names import prepare_tool_names
+    req = request()
+    req.tools[0].name = name
+    wire, _ = prepare_tool_names(req)
+    alias = wire.tools[0].name
+
+    def handler(_):
+        return httpx.Response(200, content=sse(
+            {"choices": [{"delta": {"tool_calls": [{"index": 0, "id": "call_1",
+                "function": {"name": alias[:3], "arguments": ""}}]}}]},
+            {"choices": [{"delta": {"tool_calls": [{"index": 0,
+                "function": {"name": alias[3:], "arguments": '{"query":"x"}'}}]},
+                "finish_reason": "tool_calls"}]},
+            {"type": "[DONE]"},
+        ))
+
+    events = asyncio.run(collect(provider("compatible", handler).stream(req)))
+    assert [e.data["name"] for e in events if e.event == E.tool_call_start] == [name]
+    assert json.loads("".join(e.data["arguments_delta"] for e in events
+                             if e.event == E.tool_call_delta)) == {"query": "x"}
+    assert events[-1].data["status"] == "completed"
+
+
+def sse(*events):
+    return "".join(
+        f"event: {event.get('type', 'message')}\r\ndata: {json.dumps(event, ensure_ascii=False)}\r\n\r\n"
+        for event in events
+    ).encode()
+
+
+def wire(protocol, *events):
+    if protocol == "ollama":
+        return ("\n".join(json.dumps(event, ensure_ascii=False) for event in events) + "\n").encode()
+    return sse(*events)
+
+
+def start(protocol):
+    if protocol == "responses":
+        return [{"type": "response.output_text.delta", "delta": "你好"}]
+    if protocol == "anthropic":
+        return [{"type": "message_start", "message": {"usage": {"input_tokens": 7, "output_tokens": 0}}},
+                {"type": "content_block_start", "index": 0, "content_block": {"type": "text", "text": ""}},
+                {"type": "content_block_delta", "index": 0, "delta": {"type": "text_delta", "text": "你好"}}]
+    if protocol == "compatible":
+        return [{"choices": [{"delta": {"content": "你好"}}]}]
+    return [{"message": {"content": "你好"}, "done": False}]
+
+
+def terminal(protocol):
+    if protocol == "responses":
+        return [{"type": "response.completed", "response": {"status": "completed", "usage": {"input_tokens": 7, "output_tokens": 2}}}]
+    if protocol == "anthropic":
+        return [{"type": "content_block_stop", "index": 0},
+                {"type": "message_delta", "delta": {"stop_reason": "end_turn"}, "usage": {"output_tokens": 2}},
+                {"type": "message_stop"}]
+    if protocol == "compatible":
+        return [{"choices": [{"delta": {}, "finish_reason": "stop"}], "usage": {"prompt_tokens": 7, "completion_tokens": 2}}]
+    return [{"message": {}, "done": True, "prompt_eval_count": 7, "eval_count": 2}]
+
+
+def assert_events(events):
+    assert events[-1].event == E.done
+    assert events[-1].data["status"] == ("failed" if any(event.event == E.error for event in events) else "completed")
+    assert sum(event.event == E.done for event in events) == 1
+    assert [event.sequence for event in events] == list(range(len(events)))
+    assert all(event.timestamp.tzinfo is not None for event in events)
+
+
+def assert_error(events, code):
+    assert_events(events)
+    assert events[-2].event == E.error
+    assert events[-2].data["code"] == code
+    assert SECRET not in str(events[-2].data)
+
+
+@pytest.mark.parametrize("protocol", NATIVE)
+def test_native_completion_and_history(protocol):
+    captured = {}
+
+    def handler(req):
+        captured.update(json.loads(req.content))
+        assert req.url.path == ("/v1/responses" if protocol == "responses" else "/v1/messages")
+        if protocol == "responses":
+            assert req.headers["authorization"] == f"Bearer {SECRET}"
+            body = {"status": "completed", "output": [
+                {"type": "reasoning", "summary": [{"type": "summary_text", "text": "thinking"}]},
+                {"type": "message", "content": [{"type": "output_text", "text": "完成"}]},
+                {"type": "function_call", "call_id": "next", "name": "lookup", "arguments": '{"query":"c"}'},
+            ], "usage": {"input_tokens": 10, "output_tokens": 3}}
+        else:
+            assert "authorization" not in req.headers
+            assert req.headers["x-api-key"] == SECRET
+            assert req.headers["anthropic-version"] == "2023-06-01"
+            body = {"type": "message", "content": [
+                {"type": "thinking", "thinking": "thinking", "signature": "sig"},
+                {"type": "text", "text": "完成"},
+                {"type": "tool_use", "id": "next", "name": "lookup", "input": {"query": "c"}},
+            ], "usage": {"input_tokens": 5, "cache_creation_input_tokens": 2, "cache_read_input_tokens": 3, "output_tokens": 3}}
+        return httpx.Response(200, json=body)
+
+    turn = asyncio.run(provider(protocol, handler).complete(request(history=True)))
+    assert turn.text == "完成"
+    assert (turn.input_tokens, turn.output_tokens) == (10, 3)
+    assert turn.tool_calls[0].tool_call_id == "next"
+    assert turn.tool_calls[0].arguments == {"query": "c"}
+    assert captured["stream"] is False
+    assert captured["temperature"] == 0
+    if protocol == "responses":
+        assert captured["instructions"] == "System rules"
+        assert captured["max_output_tokens"] == 512
+        assert captured["tools"][0]["parameters"] == {"type": "object"}
+        calls = [item for item in captured["input"] if item.get("type") == "function_call"]
+        outputs = [item for item in captured["input"] if item.get("type") == "function_call_output"]
+        assert [call["call_id"] for call in calls] == ["old_1", "old_2"]
+        assert json.loads(calls[1]["arguments"]) == {"query": "b"}
+        assert outputs == [{"type": "function_call_output", "call_id": "old_1", "output": '{"found":1}'},
+                           {"type": "function_call_output", "call_id": "old_2", "output": '{"found":2}'}]
+        assert {"role": "system", "content": "Additional rules"} in captured["input"]
+    else:
+        assert captured["system"] == "System rules\n\nAdditional rules"
+        assert captured["max_tokens"] == 512
+        assert captured["tools"][0]["input_schema"] == {"type": "object"}
+        assert captured["messages"][1]["content"][2] == {
+            "type": "tool_use", "id": "old_2", "name": "lookup", "input": {"query": "b"},
+        }
+        assert captured["messages"][-1] == {"role": "user", "content": [
+            {"type": "tool_result", "tool_use_id": "old_1", "content": '{"found":1}'},
+            {"type": "tool_result", "tool_use_id": "old_2", "content": '{"found":2}'},
+        ]}
+
+
+def responses_tool_events():
+    events = [
+        {"type": "response.created", "response": {"usage": {"input_tokens": 10, "output_tokens": 0}}},
+        {"type": "response.reasoning_summary_text.delta", "delta": "计划"},
+        {"type": "response.output_text.delta", "delta": "查"},
+        {"type": "response.output_text.delta", "delta": "找"},
+    ]
+    for index in (2, 3):
+        events.append({"type": "response.output_item.added", "output_index": index, "item": {
+            "id": f"item_{index}", "type": "function_call", "call_id": f"call_{index}", "name": "lookup", "arguments": "",
+        }})
+    for index, fragment in [(2, '{"query":'), (3, '{}'), (2, '"笔记"}')]:
+        events.append({"type": "response.function_call_arguments.delta", "output_index": index,
+                       "item_id": f"item_{index}", "delta": fragment})
+    for index, arguments in [(3, '{}'), (2, '{"query":"笔记"}')]:
+        events += [
+            {"type": "response.function_call_arguments.done", "output_index": index, "item_id": f"item_{index}", "arguments": arguments},
+            {"type": "response.output_item.done", "output_index": index, "item": {
+                "id": f"item_{index}", "type": "function_call", "call_id": f"call_{index}", "name": "lookup", "arguments": arguments,
+            }},
+        ]
+    events += [{"type": "future.event"}, {"type": "response.completed", "response": {
+        "status": "completed", "usage": {"input_tokens": 10, "output_tokens": 9},
+    }}]
+    return events
+
+
+def anthropic_tool_events():
+    events = [
+        {"type": "message_start", "message": {"usage": {
+            "input_tokens": 5, "cache_read_input_tokens": 3, "cache_creation_input_tokens": 2, "output_tokens": 1,
+        }}},
+        {"type": "ping"},
+        {"type": "content_block_start", "index": 0, "content_block": {"type": "thinking", "thinking": ""}},
+        {"type": "content_block_delta", "index": 0, "delta": {"type": "thinking_delta", "thinking": "计划"}},
+        {"type": "content_block_delta", "index": 0, "delta": {"type": "signature_delta", "signature": "sig"}},
+        {"type": "content_block_stop", "index": 0},
+        {"type": "content_block_start", "index": 1, "content_block": {"type": "text", "text": "查"}},
+        {"type": "content_block_delta", "index": 1, "delta": {"type": "text_delta", "text": "找"}},
+        {"type": "content_block_stop", "index": 1},
+    ]
+    for index, fragments in [(2, ['{"query":', '"笔记"}']), (3, [])]:
+        events.append({"type": "content_block_start", "index": index, "content_block": {
+            "type": "tool_use", "id": f"call_{index}", "name": "lookup", "input": {},
+        }})
+        for fragment in fragments:
+            events.append({"type": "content_block_delta", "index": index,
+                           "delta": {"type": "input_json_delta", "partial_json": fragment}})
+        events.append({"type": "content_block_stop", "index": index})
+    events += [
+        {"type": "message_delta", "delta": {"stop_reason": "tool_use"}, "usage": {"output_tokens": 4}},
+        {"type": "future.event"},
+        {"type": "message_delta", "delta": {}, "usage": {"output_tokens": 9}},
+        {"type": "message_stop"},
+    ]
+    return events
+
+
+@pytest.mark.parametrize("protocol", NATIVE)
+def test_native_stream_tools_reasoning_usage_and_fragmented_utf8(protocol):
+    frames = responses_tool_events() if protocol == "responses" else anthropic_tool_events()
+    body = Bytes(b": comment\r\n\r\n" + sse(*frames) + b"data: malformed after completion\n\n", fragment=1)
+
+    def handler(req):
+        payload = json.loads(req.content)
+        assert payload["stream"] is True
+        assert payload["tools"]
+        assert (payload.get("input") or payload.get("messages"))
+        return httpx.Response(200, stream=body)
+
+    events = asyncio.run(collect(provider(protocol, handler).stream(request(history=True))))
+    assert_events(events)
+    assert not any(event.event == E.error for event in events)
+    assert [event.data["text"] for event in events if event.event == E.text_delta] == ["查", "找"]
+    assert [event.data["text"] for event in events if event.event == E.thinking_delta] == ["计划"]
+    assert [event.data["tool_call_id"] for event in events if event.event == E.tool_call_start] == ["call_2", "call_3"]
+    assert sorted(event.data["tool_call_id"] for event in events if event.event == E.tool_call_end) == ["call_2", "call_3"]
+    for call_id, expected in [("call_2", {"query": "笔记"}), ("call_3", {})]:
+        arguments = "".join(event.data["arguments_delta"] for event in events
+                            if event.event == E.tool_call_delta and event.data["tool_call_id"] == call_id)
+        assert json.loads(arguments) == expected
+    usages = [event.data for event in events if event.event == E.usage]
+    assert usages[-1] == {"input_tokens": 10, "output_tokens": 9, "total_tokens": 19}
+    assert all(usage["input_tokens"] == 10 for usage in usages)
+    if protocol == "anthropic":
+        assert [usage["output_tokens"] for usage in usages] == [1, 4, 9]
+    assert body.closed
+
+
+@pytest.mark.parametrize("protocol", PROTOCOLS)
+def test_stream_terminal_usage_and_closure(protocol):
+    body = Bytes(wire(protocol, *start(protocol), *terminal(protocol)))
+    events = asyncio.run(collect(provider(protocol, lambda _: httpx.Response(200, stream=body)).stream(request())))
+    assert_events(events)
+    assert not any(event.event == E.error for event in events)
+    assert [event.data["text"] for event in events if event.event == E.text_delta] == ["你好"]
+    assert [event.data for event in events if event.event == E.usage][-1] == {"input_tokens": 7, "output_tokens": 2, "total_tokens": 9}
+    assert body.closed
+
+
+@pytest.mark.parametrize("protocol", PROTOCOLS)
+@pytest.mark.parametrize("empty", [False, True])
+def test_truncated_stream(protocol, empty):
+    body = Bytes(b"" if empty else wire(protocol, *start(protocol)))
+    events = asyncio.run(collect(provider(protocol, lambda _: httpx.Response(200, stream=body)).stream(request())))
+    assert_error(events, "PROVIDER_STREAM_TRUNCATED")
+    assert body.closed
+
+
+@pytest.mark.parametrize("protocol", PROTOCOLS)
+@pytest.mark.parametrize("bad", [b"not-json", b"[]", b"null", b'{"usage":'])
+def test_malformed_stream_is_sanitized(protocol, bad):
+    suffix = bad + b"\n" if protocol == "ollama" else b"data: " + bad + b"\n\n"
+    body = Bytes(wire(protocol, *start(protocol)) + suffix)
+    events = asyncio.run(collect(provider(protocol, lambda _: httpx.Response(200, stream=body)).stream(request())))
+    assert_error(events, "PROVIDER_INVALID_RESPONSE")
+    assert body.closed
+
+
+@pytest.mark.parametrize("protocol", PROTOCOLS)
+@pytest.mark.parametrize("error_type,code", [("rate_limit_error", "PROVIDER_RATE_LIMITED"),
+                                           ("authentication_error", "PROVIDER_AUTH_FAILED"),
+                                           ("overloaded_error", "PROVIDER_UNAVAILABLE")])
+def test_in_band_error_after_partial_output(protocol, error_type, code):
+    body = Bytes(wire(protocol, *start(protocol), {"type": "error", "error": {"type": error_type, "message": SECRET}}))
+    events = asyncio.run(collect(provider(protocol, lambda _: httpx.Response(200, stream=body)).stream(request())))
+    assert any(event.event == E.text_delta for event in events)
+    assert_error(events, code)
+    assert body.closed
+
+
+@pytest.mark.parametrize("protocol", PROTOCOLS)
+@pytest.mark.parametrize("status,code", [(400, "PROVIDER_INVALID_REQUEST"), (401, "PROVIDER_AUTH_FAILED"),
+                                       (403, "PROVIDER_AUTH_FAILED"), (404, "MODEL_NOT_FOUND"),
+                                       (429, "PROVIDER_RATE_LIMITED"), (500, "PROVIDER_UNAVAILABLE")])
+def test_http_errors_completion_and_stream(protocol, status, code):
+    adapter = provider(protocol, lambda _: httpx.Response(status, text=SECRET))
+    with pytest.raises(ProviderError) as exc:
+        asyncio.run(adapter.complete(request()))
+    assert exc.value.code == code
+    assert SECRET not in str(exc.value)
+    assert_error(asyncio.run(collect(adapter.stream(request()))), code)
+
+
+@pytest.mark.parametrize("protocol", PROTOCOLS)
+@pytest.mark.parametrize("body,code", [(b"broken", "PROVIDER_INVALID_RESPONSE"),
+                                     (b"[]", "PROVIDER_INVALID_RESPONSE"),
+                                     (b"{}", "PROVIDER_INVALID_RESPONSE"),
+                                     (json.dumps({"error": {"code": "invalid_api_key", "message": SECRET}}).encode(), "PROVIDER_AUTH_FAILED")])
+def test_bad_completion(protocol, body, code):
+    adapter = provider(protocol, lambda _: httpx.Response(200, content=body))
+    with pytest.raises(ProviderError) as exc:
+        asyncio.run(adapter.complete(request()))
+    assert exc.value.code == code
+    assert SECRET not in str(exc.value)
+
+
+@pytest.mark.parametrize("protocol", PROTOCOLS)
+@pytest.mark.parametrize("error,code", [(httpx.ReadTimeout, "PROVIDER_TIMEOUT"),
+                                      (httpx.ConnectError, "PROVIDER_UNAVAILABLE")])
+def test_transport_error_mapping(protocol, error, code):
+    def handler(req):
+        raise error(SECRET, request=req)
+
+    adapter = provider(protocol, handler)
+    with pytest.raises(ProviderError) as exc:
+        asyncio.run(adapter.complete(request()))
+    assert exc.value.code == code
+    assert SECRET not in str(exc.value)
+    assert_error(asyncio.run(collect(adapter.stream(request()))), code)
+
+
+@pytest.mark.parametrize("protocol", PROTOCOLS)
+@pytest.mark.parametrize("cancel", [True, False])
+def test_incremental_delivery_cancellation_and_explicit_close(protocol, cancel):
+    async def scenario():
+        body = GatedBytes(wire(protocol, *start(protocol)))
+        adapter = provider(protocol, lambda _: httpx.Response(200, stream=body))
+        iterator = adapter.stream(request())
+        seen = []
+        while True:
+            event = await asyncio.wait_for(anext(iterator), timeout=1)
+            seen.append(event)
+            if event.event == E.text_delta:
+                break
+        # The first token arrives while the response is still open and blocked.
+        assert seen[-1].data["text"] == "你好"
+        assert not body.closed
+        if cancel:
+            pending = asyncio.create_task(anext(iterator))
+            await asyncio.wait_for(body.waiting.wait(), timeout=1)
+            pending.cancel()
+            with pytest.raises(asyncio.CancelledError):
+                await pending
+        else:
+            await iterator.aclose()
+        assert body.closed
+        assert not any(event.event in {E.error, E.done} for event in seen)
+
+    asyncio.run(scenario())
+
+
+@pytest.mark.parametrize("protocol", NATIVE)
+def test_cancellation_before_response_headers(protocol):
+    async def scenario():
+        entered = asyncio.Event()
+        closed = asyncio.Event()
+
+        async def handler(req):
+            entered.set()
+            try:
+                await asyncio.Event().wait()
+            finally:
+                closed.set()
+
+        adapter = provider(protocol, handler)
+        pending = asyncio.create_task(adapter.complete(request()))
+        await asyncio.wait_for(entered.wait(), timeout=1)
+        pending.cancel()
+        with pytest.raises(asyncio.CancelledError):
+            await pending
+        assert closed.is_set()
+
+    asyncio.run(scenario())
+
+
+@pytest.mark.parametrize("protocol", NATIVE)
+def test_native_discovery_does_not_claim_non_chat_capabilities(protocol):
+    def handler(req):
+        assert req.url.path == "/v1/models"
+        return httpx.Response(200, json={"data": [{"id": name} for name in ["chat-model", "text-embedding-3-small", "whisper-1", "gpt-audio"]]})
+
+    models = asyncio.run(provider(protocol, handler).list_models())
+    assert ModelCapability.chat in models[0].capabilities
+    assert models[1].capabilities == [ModelCapability.embedding]
+    assert all(ModelCapability.chat not in model.capabilities for model in models[1:])
+
+
+@pytest.mark.parametrize("protocol", NATIVE)
+def test_native_structured_format_mapping(protocol):
+    adapter = provider(protocol, lambda _: pytest.fail("No network expected"))
+    req = request()
+    req.response_format = {"type": "json_schema", "json_schema": {
+        "name": "answer", "strict": True, "schema": {"type": "object", "properties": {}},
+    }}
+    payload = adapter._payload(req, stream=False)
+    format_ = payload["text"]["format"] if protocol == "responses" else payload["output_config"]["format"]
+    assert format_["type"] == "json_schema"
+    assert format_["schema"] == {"type": "object", "properties": {}}
+    if protocol == "responses":
+        assert format_["name"] == "answer"
+        assert format_["strict"] is True
+
+
+@pytest.mark.parametrize("protocol", NATIVE)
+def test_invalid_tool_arguments_and_unclosed_tool(protocol):
+    frames = responses_tool_events() if protocol == "responses" else anthropic_tool_events()
+    # A syntactically valid terminal cannot rescue an unfinished tool block.
+    index = next(i for i, frame in enumerate(frames)
+                 if frame["type"] in {"response.function_call_arguments.delta", "content_block_delta"}
+                 and (frame.get("output_index") == 2 or frame.get("index") == 2))
+    partial = frames[:index + 1]
+    final = frames[-1]
+    events = asyncio.run(collect(provider(protocol, lambda _: httpx.Response(200, content=sse(*partial, final))).stream(request())))
+    assert_error(events, "PROVIDER_STREAM_TRUNCATED")
+    assert not any(event.event == E.tool_call_end for event in events)
+
+    for frame in frames:
+        if frame["type"] == "response.function_call_arguments.done":
+            frame["arguments"] = "[]"
+            break
+        if frame["type"] == "content_block_delta" and frame.get("index") == 2:
+            frame["delta"]["partial_json"] = "malformed"
+            break
+    events = asyncio.run(collect(provider(protocol, lambda _: httpx.Response(200, content=sse(*frames))).stream(request())))
+    assert_error(events, "PROVIDER_INVALID_RESPONSE")
+
+
+@pytest.mark.parametrize("kind,code", [("response.failed", "PROVIDER_UNAVAILABLE"),
+                                     ("response.incomplete", "PROVIDER_INCOMPLETE_RESPONSE")])
+def test_responses_failed_and_incomplete(kind, code):
+    frame = {"type": kind, "response": {"status": kind.split(".")[1], "incomplete_details": {"reason": SECRET}}}
+    events = asyncio.run(collect(provider("responses", lambda _: httpx.Response(200, content=sse(*start("responses"), frame))).stream(request())))
+    assert_error(events, code)
+
+
+def test_sse_multiline_data_and_event_name_without_json_type():
+    body = (b': keepalive\n\nevent: response.output_text.delta\ndata: {\ndata: "delta": "hello"\ndata: }\n\n'
+            + sse({"type": "response.completed", "response": {"status": "completed"}}))
+    events = asyncio.run(collect(provider("responses", lambda _: httpx.Response(200, content=body)).stream(request())))
+    assert_events(events)
+    assert [event.data["text"] for event in events if event.event == E.text_delta] == ["hello"]
+    assert not any(event.event == E.error for event in events)
+
+
+def test_ollama_history_options_and_in_band_string_error():
+    captured = {}
+
+    def handler(req):
+        captured.update(json.loads(req.content))
+        return httpx.Response(200, json={"error": SECRET})
+
+    with pytest.raises(ProviderError) as exc:
+        asyncio.run(provider("ollama", handler).complete(request(history=True)))
+    assert exc.value.code == "PROVIDER_UNAVAILABLE"
+    assert SECRET not in str(exc.value)
+    assert captured["messages"][-1]["tool_name"] == "lookup"
+    assert captured["options"] == {"temperature": 0.0, "num_predict": 512}
+
+@pytest.mark.parametrize("protocol", PROTOCOLS)
+@pytest.mark.parametrize("streaming", [False, True])
+def test_namespaced_tools_roundtrip_without_changing_internal_request(protocol, streaming):
+    import re
+    model_request = request(history=True)
+    original_name = "mcp.my-server.search.notes"
+    model_request.tools[0].name = original_name
+    for message in model_request.messages:
+        for call in message.tool_calls:
+            call.name = original_name
+    before = model_request.model_dump()
+
+    def handler(req):
+        payload = json.loads(req.content)
+        definition = payload["tools"][0]
+        name = (definition.get("function") or definition)["name"]
+        assert name != original_name and re.fullmatch(r"[a-zA-Z0-9_-]{1,64}", name)
+        assert original_name not in req.content.decode()
+        if protocol == "responses":
+            item = {"type": "function_call", "id": "item1", "call_id": "call1", "name": name, "arguments": "{}"}
+            body = {"status": "completed", "output": [item]}
+            events = [
+                {"type": "response.output_item.done", "output_index": 0, "item": item},
+                {"type": "response.completed", "response": {"status": "completed"}},
+            ]
+        elif protocol == "anthropic":
+            item = {"type": "tool_use", "id": "call1", "name": name, "input": {}}
+            body = {"content": [item]}
+            events = [
+                {"type": "message_start", "message": {}},
+                {"type": "content_block_start", "index": 0, "content_block": item},
+                {"type": "content_block_stop", "index": 0},
+                {"type": "message_stop"},
+            ]
+        elif protocol == "compatible":
+            item = {"id": "call1", "function": {"name": name, "arguments": "{}"}}
+            body = {"choices": [{"message": {"tool_calls": [item]}}]}
+            events = [{"choices": [{"delta": {"tool_calls": [{"index": 0, **item}]}, "finish_reason": "tool_calls"}]}]
+        else:
+            item = {"function": {"name": name, "arguments": {}}}
+            body = {"message": {"tool_calls": [item]}, "done": True}
+            events = [body]
+        return httpx.Response(200, content=wire(protocol, *events)) if streaming else httpx.Response(200, json=body)
+
+    adapter = provider(protocol, handler)
+    if streaming:
+        events = asyncio.run(collect(adapter.stream(model_request)))
+        assert_events(events)
+        assert [event.data["name"] for event in events if event.event == E.tool_call_start] == [original_name]
+    else:
+        assert asyncio.run(adapter.complete(model_request)).tool_calls[0].name == original_name
+    assert model_request.model_dump() == before
+
+def test_chat_route_closes_upstream_and_sanitizes_unexpected_errors(monkeypatch):
+    from types import SimpleNamespace
+    from datetime import datetime, timezone
+    from app import routes
+    from app.contracts import ChatRequest, ModelEvent
+    closed = []
+
+    class Adapter:
+        async def stream(self, request):
+            try:
+                yield ModelEvent(event=E.text_delta, sequence=0, data={"text": "first"}, timestamp=datetime.now(timezone.utc))
+                raise RuntimeError(SECRET)
+            finally:
+                closed.append(True)
+
+    monkeypatch.setattr(routes, "provider_or_404", lambda _: SimpleNamespace(adapter=Adapter()))
+
+    async def scenario():
+        response = await routes.chat(ChatRequest(provider_id="test", model="test", messages=[]))
+        iterator = response.body_iterator
+        await anext(iterator)
+        await iterator.aclose()
+        assert len(closed) == 1
+        response = await routes.chat(ChatRequest(provider_id="test", model="test", messages=[]))
+        items = [json.loads(chunk.split("data: ")[1].strip()) async for chunk in response.body_iterator]
+        assert [item["sequence"] for item in items] == [0, 1, 2]
+        assert items[-1]["data"]["status"] == "failed"
+        assert SECRET not in str(items)
+        assert len(closed) == 2
+
+    asyncio.run(scenario())

backend/tests/test_retrieval.py

@@ -436,6 +436,83 @@ def test_fts_pagination_is_not_truncated_at_one_thousand(vault) -> None:
     assert len(response.items) == 10
 
 
+def test_fts_score_threshold_filters_before_total(vault) -> None:
+    """score_threshold 先于计数与分页生效：total 反映过滤后数量，与 items 一致。
+
+    高阈值过滤掉全部结果时 total==0 且 items 为空，杜绝「空页但 total>0」的
+    不一致（审阅 P2-7）。
+    """
+    from app.retrieval.engine import engine
+    from app.services import note_service
+
+    # 10 个 block，含「目标」次数递增，bm25 分数各异，min-max 归一化后分数落在 [0,1]
+    markdown = "\n\n".join(f"{'目标' * i} 分隔内容" for i in range(1, 11))
+    asyncio.run(
+        note_service.create_note(title="阈值过滤", markdown=markdown, folder="", tags=[])
+    )
+
+    all_hits = asyncio.run(
+        engine.search(
+            SearchRequest(query="目标", mode=SearchMode.fts, limit=20, score_threshold=0.0)
+        )
+    )
+    filtered = asyncio.run(
+        engine.search(
+            SearchRequest(query="目标", mode=SearchMode.fts, limit=20, score_threshold=0.5)
+        )
+    )
+    none = asyncio.run(
+        engine.search(
+            SearchRequest(query="目标", mode=SearchMode.fts, limit=20, score_threshold=2.0)
+        )
+    )
+
+    assert all_hits.page.total >= 10
+    assert 0 < filtered.page.total < all_hits.page.total  # 阈值过滤掉部分而非全部
+    assert filtered.page.total == len(filtered.items)
+    assert none.page.total == 0
+    assert none.items == []
+
+
+def test_fts_offset_beyond_end_reports_real_total(vault) -> None:
+    """offset 越过末页时 items 为空，但 total 仍为真实命中数而非归零。"""
+    from app.retrieval.engine import engine
+    from app.services import note_service
+
+    asyncio.run(
+        note_service.create_note(title="越界分页", markdown="检索 检索 检索 检索", folder="", tags=[])
+    )
+
+    resp = asyncio.run(
+        engine.search(SearchRequest(query="检索", mode=SearchMode.fts, limit=10, offset=100))
+    )
+    assert resp.page.total >= 1
+    assert resp.items == []
+
+
+def test_fts_not_truncated_at_five_thousand(vault) -> None:
+    """FTS 结果不再被 5000 条上限截断：>5000 命中时 total 为真实计数，末页仍可访问。"""
+    from app.retrieval.engine import engine
+    from app.services import note_service
+
+    markdown = "\n\n".join(f"共同词 q{i}" for i in range(5010))
+    asyncio.run(
+        note_service.create_note(title="五千条分页", markdown=markdown, folder="", tags=[])
+    )
+
+    first = asyncio.run(
+        engine.search(SearchRequest(query="共同词", mode=SearchMode.fts, limit=10, offset=0))
+    )
+    assert first.page.total == 5010
+    assert len(first.items) == 10
+
+    last = asyncio.run(
+        engine.search(SearchRequest(query="共同词", mode=SearchMode.fts, limit=10, offset=5005))
+    )
+    assert last.page.total == 5010
+    assert len(last.items) == 5
+
+
 # --------------------------------------------------------------------------- #
 # 审阅回归：PATCH tags 语义 / 向量-块一致性 / 过滤漏召回 / rebuild 语义与回滚
 # --------------------------------------------------------------------------- #
@@ -563,9 +640,10 @@ def test_rebuild_failure_restores_old_index(vault, monkeypatch) -> None:
     assert repository.stats() == before  # 旧索引已恢复，无半成品
 
 
-def test_first_rebuild_failure_removes_partial_database(vault, monkeypatch) -> None:
+def test_first_rebuild_failure_leaves_no_partial_index(vault, monkeypatch) -> None:
     """首次启动没有旧库时，失败也不能留下已经写入的部分索引。"""
     from app.services import index_service
+    from app import repository
 
     _write_vault(
         vault,
@@ -574,17 +652,17 @@ def test_first_rebuild_failure_removes_partial_database(vault, monkeypatch) -> N
     real_index = index_service.index_note
     calls = {"count": 0}
 
-    async def fail_on_second(parsed):
+    async def fail_on_second(parsed, **kwargs):
         calls["count"] += 1
         if calls["count"] == 2:
             raise RuntimeError("injected first-rebuild failure")
-        await real_index(parsed)
+        await real_index(parsed, **kwargs)
 
     monkeypatch.setattr(index_service, "index_note", fail_on_second)
     with pytest.raises(RuntimeError):
         asyncio.run(index_service.rebuild(IndexRebuildRequest(scope="all")))
 
-    assert not get_settings().db_path.exists()
+    assert repository.stats() == {"notes": 0, "blocks": 0}
 
 
 def test_rebuild_preserves_task_note_links(vault) -> None:

backend/tests/test_routed_retrieval.py

@@ -0,0 +1,466 @@
+"""Phase E route integration: deterministic runtimes, isolated DBs, no network."""
+
+from __future__ import annotations
+
+import asyncio
+import json
+from dataclasses import dataclass, field
+from types import SimpleNamespace
+
+import pytest
+
+from app import repository
+from app.config import get_settings
+from app.contracts import IndexRebuildRequest, SearchMode, SearchRequest
+from app.database.db import connect, transaction
+from app.retrieval import routed_vectors
+from app.retrieval.embedding import HashEmbeddingProvider
+from app.retrieval.engine import RetrievalEngine, engine
+from app.retrieval.reranker import LexicalReranker
+from app.retrieval.vectorstore import SqliteVecStore, VectorHit
+from app.services import index_service, note_service
+
+
+@dataclass
+class FakeRuntime:
+    model_id: str = "space-a"
+    dimensions: int = 3  # Deliberately differs from sqlite-vec's fixed 128.
+    source: str = "api"
+    error: BaseException | None = None
+    calls: list[list[str]] = field(default_factory=list)
+    result_override: object | None = None
+
+    async def embed(self, texts):
+        self.calls.append(list(texts))
+        if self.error is not None:
+            raise self.error
+        if self.result_override is not None:
+            return self.result_override
+        vectors = []
+        for text in texts:
+            # The API associates "apple" with banana; hash retrieval picks apple.
+            first = text == "apple orchard"
+            if self.model_id == "space-b":
+                first = not first
+            vectors.append(([1.0, 0.0] if first else [0.0, 1.0]) + [0.0] * (self.dimensions - 2))
+        return SimpleNamespace(
+            vectors=vectors, source=self.source, model_id=self.model_id,
+            dimensions=self.dimensions, fallback_reason=None,
+        )
+
+
+@pytest.fixture
+def runtime(monkeypatch):
+    runtime = FakeRuntime()
+    monkeypatch.setattr(routed_vectors, "get_model_routing", lambda: runtime)
+    return runtime
+
+
+async def seed():
+    apple = await note_service.create_note(
+        title="Apple", markdown="apple orchard", folder=None, tags=[],
+    )
+    banana = await note_service.create_note(
+        title="Banana", markdown="banana grove", folder=None, tags=[],
+    )
+    return apple, banana
+
+
+@pytest.mark.parametrize("outcome", ["api", "api_failure", "missing_space"])
+def test_benchmark_reports_actual_embedding_and_fallback(runtime, outcome):
+    from app.benchmarks import service
+    from app.contracts import RAGRunRequest
+
+    async def scenario():
+        apple, banana = await seed()
+        if outcome == "api_failure":
+            runtime.result_override = SimpleNamespace(source="local", fallback_reason="PROVIDER_TIMEOUT")
+        elif outcome == "missing_space":
+            runtime.model_id = "space-without-index"
+        directory = get_settings().benchmark_datasets_path
+        directory.mkdir(parents=True, exist_ok=True)
+        (directory / "routing.json").write_text(json.dumps({
+            "dataset_id": "routing", "kind": "rag", "version": "1",
+            "cases": [{"case_id": "query", "query": "apple", "expected_note_ids": [banana.note_id]}],
+        }), encoding="utf-8")
+        run = await service.create_rag_run(RAGRunRequest(
+            dataset_id="routing", modes=[SearchMode.fts, SearchMode.vector],
+        ))
+        await service.wait_for_run(run.run_id)
+        report = service.get_report(run.run_id)
+        assert report.config_snapshot["embedding"]["policy"] == "per_case"
+        fts, vector = report.cases
+        assert fts.embedding == {"source": "not_used"}
+        if outcome == "api":
+            assert vector.embedding["source"] == "api"
+            assert vector.embedding["model_id"] == "space-a"
+            assert vector.embedding["dimensions"] == 3
+            assert vector.retrieved_note_ids[0] == banana.note_id
+        else:
+            assert vector.embedding["source"] == "local"
+            assert vector.embedding["model_id"] == "hash-v1"
+            assert vector.embedding["dimensions"] == 128
+            assert vector.retrieved_note_ids[0] == apple.note_id
+            if outcome == "api_failure":
+                assert vector.embedding["fallback_reason"] == "PROVIDER_TIMEOUT"
+            if outcome == "missing_space":
+                assert vector.embedding["fallback_reason"] == "REMOTE_INDEX_UNAVAILABLE"
+                assert vector.embedding["attempted_space"]["model_id"] == "space-without-index"
+        events = service.get_events(run.run_id)
+        case_events = [e for e in events if e.event.value == "CaseCompleted"]
+        assert case_events[-1].data["embedding"] == vector.embedding
+
+    asyncio.run(scenario())
+
+
+def test_embedding_observations_are_isolated_between_concurrent_searches(runtime, monkeypatch):
+    from app.retrieval.provenance import capture_embedding
+
+    async def scenario():
+        await seed()
+        original = runtime.embed
+
+        async def embed(texts):
+            await asyncio.sleep(0)
+            if texts == ["offline"]:
+                raise RuntimeError("private upstream details")
+            return await original(texts)
+
+        monkeypatch.setattr(runtime, "embed", embed)
+
+        async def query(text):
+            with capture_embedding() as observation:
+                await engine.search(SearchRequest(query=text, mode=SearchMode.vector))
+            return observation
+
+        remote, local, another = await asyncio.gather(query("apple"), query("offline"), query("apple"))
+        assert remote["source"] == another["source"] == "api"
+        assert local["source"] == "local"
+        assert local["fallback_reason"] == "REMOTE_EMBEDDING_UNAVAILABLE"
+        assert "fallback_reason" not in remote or remote["fallback_reason"] is None
+        assert "private upstream" not in json.dumps(local)
+
+    asyncio.run(scenario())
+
+
+@pytest.mark.parametrize("failure", ["cancel", "write"])
+def test_rebuild_failure_preserves_concurrent_configuration_and_all_indexes(runtime, monkeypatch, failure):
+    from app.container import container
+    from app.contracts import ModelRoutingConfig, ProviderConfig, ProviderType
+    from app.services import task_service
+
+    async def scenario():
+        apple, _ = await seed()
+        task = task_service.create_task(title="before", note_id=apple.note_id)
+        before = {table: [tuple(row) for row in rows(f"SELECT * FROM {table}")]
+                  for table in ("notes", "blocks", "blocks_fts", "vec_blocks", "index_meta", "routed_block_vectors")}
+        container.model_routing.update(ModelRoutingConfig())
+        entered, release = asyncio.Event(), asyncio.Event()
+        original_embed = runtime.embed
+
+        async def pending_embed(texts):
+            entered.set()
+            await release.wait()
+            return await original_embed(texts)
+
+        monkeypatch.setattr(runtime, "embed", pending_embed)
+        original_index = index_service.index_note
+        writes = 0
+
+        async def fail_write(parsed, **kwargs):
+            nonlocal writes
+            await original_index(parsed, **kwargs)
+            writes += 1
+            if writes == 2:
+                raise RuntimeError("injected write failure")
+
+        if failure == "write":
+            monkeypatch.setattr(index_service, "index_note", fail_write)
+        rebuilding = asyncio.create_task(index_service.rebuild(IndexRebuildRequest()))
+        await asyncio.wait_for(entered.wait(), timeout=5)
+        saved = container.model_routing.update(container.model_routing.configuration())
+        config = ProviderConfig(provider_id="concurrent", provider_type=ProviderType.openai_compatible,
+                                name="saved during rebuild", base_url="https://unused.invalid/v1")
+        container.providers.register(config, container.provider_factory.build(config))
+        task_service.update_task(task.task_id, {"title": "saved during rebuild"})
+        # Preparation keeps the old searchable index intact while API I/O is pending.
+        assert repository.stats()["notes"] == 2
+        if failure == "cancel":
+            rebuilding.cancel()
+            expected = asyncio.CancelledError
+        else:
+            release.set()
+            expected = RuntimeError
+        with pytest.raises(expected):
+            await rebuilding
+        assert container.model_routing.configuration().version == saved.config.version
+        assert rows("SELECT provider_id FROM provider_configs")[-1][0] == "concurrent"
+        restored = task_service.get_task(task.task_id)
+        assert restored.title == "saved during rebuild"
+        assert restored.note_id == apple.note_id
+        for table, values in before.items():
+            assert [tuple(row) for row in rows(f"SELECT * FROM {table}")] == values
+
+    asyncio.run(scenario())
+
+
+def local_engine():
+    return RetrievalEngine(HashEmbeddingProvider(), LexicalReranker(), SqliteVecStore())
+
+
+def request(mode=SearchMode.vector):
+    return SearchRequest(query="apple", mode=mode, limit=10)
+
+
+def rows(sql, parameters=()):
+    conn = connect()
+    try:
+        return conn.execute(sql, parameters).fetchall()
+    finally:
+        conn.close()
+
+
+def test_api_index_and_query_use_matching_space_and_keep_local_metadata(runtime):
+    async def scenario():
+        apple, banana = await seed()
+        result = await engine.search(request())
+        assert result.items[0].note_id == banana.note_id
+        baseline = await local_engine().search(request())
+        assert baseline.items[0].note_id == apple.note_id
+        assert rows("SELECT DISTINCT space_id, dimensions FROM routed_block_vectors")[0][:] == ("space-a", 3)
+        assert rows("SELECT COUNT(*) FROM routed_block_vectors")[0][0] == len(apple.blocks) + len(banana.blocks)
+        meta = repository.get_index_meta()
+        assert meta["embedding_model"] == "hash-v1"
+        assert meta["embedding_dim"] == "128"
+        assert len(runtime.calls) == 3
+
+    asyncio.run(scenario())
+
+
+@pytest.mark.parametrize("failure", ["exception", "local", "missing", "dimension", "corrupt"])
+def test_query_falls_back_to_exact_local_results(runtime, failure):
+    async def scenario():
+        await seed()
+        if failure == "exception":
+            runtime.error = RuntimeError("offline")
+        elif failure == "local":
+            runtime.source = "local"
+        elif failure == "missing":
+            rows("DELETE FROM routed_block_vectors WHERE block_id = (SELECT MIN(block_id) FROM blocks)")
+        elif failure == "dimension":
+            runtime.dimensions = 4
+        else:
+            rows("UPDATE routed_block_vectors SET vector = ?", ("[0, 0, 0]",))
+        actual = await engine.search(request())
+        baseline = await local_engine().search(request())
+        assert actual == baseline
+
+    asyncio.run(scenario())
+
+
+def test_same_dimension_model_switch_never_combines_partial_spaces(runtime):
+    async def scenario():
+        apple, banana = await seed()
+        baseline = await local_engine().search(request())
+        runtime.model_id = "space-b"
+        assert await engine.search(request()) == baseline
+        await note_service.update_note(apple.note_id, markdown="apple orchard")
+        assert {row[0] for row in rows("SELECT DISTINCT space_id FROM routed_block_vectors")} == {"space-a", "space-b"}
+        assert await routed_vectors.search_remote("apple", top_k=10) is None
+        assert await engine.search(request()) == baseline
+        runtime.model_id = "space-a"
+        assert await engine.search(request()) == baseline
+        runtime.model_id = "space-b"
+        await note_service.update_note(banana.note_id, markdown="banana grove")
+        hits = await routed_vectors.search_remote("apple", top_k=10)
+        assert hits is not None and hits[0].id == banana.blocks[0].block_id
+        assert (await engine.search(request())).items[0].note_id == banana.note_id
+
+    asyncio.run(scenario())
+
+
+def test_complete_spaces_coexist_but_only_requested_space_is_ranked(runtime):
+    async def scenario():
+        apple, banana = await seed()
+        conn = connect()
+        try:
+            with transaction(conn):
+                routed_vectors.store_remote(
+                    conn, [apple.blocks[0].block_id, banana.blocks[0].block_id],
+                    routed_vectors.RemoteEmbeddings("space-b", 3, [[1, 0, 0], [0, 1, 0]]),
+                )
+        finally:
+            conn.close()
+        assert (await engine.search(request())).items[0].note_id == banana.note_id
+        runtime.model_id = "space-b"
+        result = await engine.search(request())
+        assert len(result.items) == 2
+        assert result.items[0].note_id == apple.note_id
+
+    asyncio.run(scenario())
+
+
+def test_failed_note_embedding_preserves_save_and_forces_coverage_fallback(runtime):
+    async def scenario():
+        apple, banana = await seed()
+        runtime.error = RuntimeError("offline")
+        await note_service.update_note(banana.note_id, markdown="banana changed")
+        assert (await note_service.get_note(banana.note_id)).markdown == "banana changed"
+        assert rows("SELECT COUNT(*) FROM routed_block_vectors")[0][0] == len(apple.blocks)
+        runtime.error = None
+        assert await engine.search(request()) == await local_engine().search(request())
+
+    asyncio.run(scenario())
+
+
+@pytest.mark.parametrize("vectors, dimensions, space", [
+    ([], 3, "space-a"),
+    ([[1, 0]], 3, "space-a"),
+    ([[0, 0, 0]], 3, "space-a"),
+    ([[float("nan"), 0, 0]], 3, "space-a"),
+    ([[float("inf"), 0, 0]], 3, "space-a"),
+    ([[True, 0, 0]], 3, "space-a"),
+    ([[1, 0, 0]], 0, "space-a"),
+    ([[1, 0, 0]], 3, "hash-v1"),
+])
+def test_invalid_remote_batch_does_not_break_note_saving(runtime, vectors, dimensions, space):
+    runtime.result_override = SimpleNamespace(
+        source="api", vectors=vectors, dimensions=dimensions, model_id=space,
+    )
+
+    async def scenario():
+        note = await note_service.create_note(title="Apple", markdown="apple orchard", folder=None, tags=[])
+        assert (await local_engine().search(request())).items[0].note_id == note.note_id
+        assert await routed_vectors.search_remote("apple", top_k=10) is None
+
+    asyncio.run(scenario())
+
+
+def test_remote_storage_failure_rolls_back_batch_but_keeps_local_index(runtime):
+    async def scenario():
+        await seed()
+        rows("""CREATE TRIGGER reject_remote_vector BEFORE INSERT ON routed_block_vectors
+                WHEN (SELECT content FROM blocks WHERE block_id = NEW.block_id) = 'second'
+                BEGIN SELECT RAISE(ABORT, 'simulated storage failure'); END""")
+        note = await note_service.create_note(
+            title="Multi", markdown="first\n\nsecond", folder=None, tags=[],
+        )
+        assert len(note.blocks) == 2
+        assert rows(
+            "SELECT COUNT(*) FROM routed_block_vectors r JOIN blocks b USING(block_id) WHERE b.note_id = ?",
+            (note.note_id,),
+        )[0][0] == 0
+        assert rows("SELECT COUNT(*) FROM vec_blocks")[0][0] == rows("SELECT COUNT(*) FROM blocks")[0][0]
+        assert (get_settings().vault_path / note.file_path).exists()
+
+    asyncio.run(scenario())
+
+
+def test_rebuild_and_delete_clear_old_remote_rows_through_foreign_keys(runtime):
+    async def scenario():
+        apple, _ = await seed()
+        await note_service.delete_note(apple.note_id)
+        assert rows("SELECT COUNT(*) FROM routed_block_vectors")[0][0] == 1
+        runtime.source = "local"
+        job = await index_service.rebuild(IndexRebuildRequest())
+        assert job.status == "completed"
+        assert rows("SELECT COUNT(*) FROM routed_block_vectors")[0][0] == 0
+        assert rows("SELECT COUNT(*) FROM vec_blocks")[0][0] == 1
+        runtime.source = "api"
+        runtime.model_id = "space-b"
+        await index_service.rebuild(IndexRebuildRequest())
+        assert [row[0] for row in rows("SELECT space_id FROM routed_block_vectors")] == ["space-b"]
+
+    asyncio.run(scenario())
+
+
+@pytest.mark.parametrize("operation", ["save", "query", "rebuild"])
+def test_cancellation_propagates_and_mutations_roll_back(runtime, operation):
+    async def scenario():
+        apple, _ = await seed()
+        before = [tuple(row) for row in rows("SELECT * FROM routed_block_vectors ORDER BY block_id")]
+        runtime.error = asyncio.CancelledError()
+        with pytest.raises(asyncio.CancelledError):
+            if operation == "query":
+                await engine.search(request())
+            elif operation == "rebuild":
+                await index_service.rebuild(IndexRebuildRequest())
+            else:
+                await note_service.update_note(apple.note_id, markdown="changed")
+        assert (await note_service.get_note(apple.note_id)).markdown == "apple orchard"
+        assert [tuple(row) for row in rows("SELECT * FROM routed_block_vectors ORDER BY block_id")] == before
+
+    asyncio.run(scenario())
+
+
+@pytest.mark.parametrize("injected", ["embedding", "vector_store", "constructor"])
+def test_injected_engine_dependencies_are_respected(runtime, monkeypatch, injected):
+    async def scenario():
+        apple, _ = await seed()
+        target = engine
+        if injected == "constructor":
+            target = local_engine()
+        elif injected == "embedding":
+            monkeypatch.setattr(engine, "embedding", HashEmbeddingProvider())
+        else:
+            class FakeStore:
+                async def search(self, vector, *, top_k):
+                    assert len(vector) == 128
+                    return [VectorHit(id=apple.blocks[0].block_id, score=1.0)]
+
+            monkeypatch.setattr(engine, "vector_store", FakeStore())
+        runtime.calls.clear()
+        assert (await target.search(request())).items[0].note_id == apple.note_id
+        assert runtime.calls == []
+
+    asyncio.run(scenario())
+
+
+def test_fts_skips_routing_and_hybrid_uses_routed_vector_channel(runtime, monkeypatch):
+    async def scenario():
+        _, banana = await seed()
+        runtime.calls.clear()
+        await engine.search(request(SearchMode.fts))
+        assert runtime.calls == []
+        # Empty lexical channel isolates the vector contribution to hybrid fusion.
+        monkeypatch.setattr(repository, "fts_search", lambda *_: [])
+
+        class PreserveOrder:
+            async def rerank(self, query, candidates):
+                return sorted(candidates, key=lambda candidate: -candidate.score)
+
+        monkeypatch.setattr(engine, "reranker", PreserveOrder())
+        result = await engine.search(request(SearchMode.hybrid))
+        assert result.items[0].note_id == banana.note_id
+        assert runtime.calls == [["apple"]]
+
+    asyncio.run(scenario())
+
+
+def test_arbitrary_dimensions_and_extreme_finite_values(runtime):
+    dimensions = 257
+    runtime.result_override = SimpleNamespace(
+        source="api", model_id="space-wide", dimensions=dimensions,
+        vectors=[[1e308, 1e308] + [0.0] * (dimensions - 2)],
+    )
+
+    async def scenario():
+        note = await note_service.create_note(title="Apple", markdown="apple orchard", folder=None, tags=[])
+        hits = await routed_vectors.search_remote("apple", top_k=1)
+        assert hits is not None and hits[0].id == note.blocks[0].block_id
+        assert hits[0].score == pytest.approx(1.0)
+        vector = json.loads(rows("SELECT vector FROM routed_block_vectors")[0][0])
+        assert len(vector) == dimensions
+
+    asyncio.run(scenario())
+
+
+def test_missing_runtime_uses_unchanged_local_retrieval(runtime, monkeypatch):
+    monkeypatch.setattr(routed_vectors, "get_model_routing", lambda: None)
+
+    async def scenario():
+        await seed()
+        assert await engine.search(request()) == await local_engine().search(request())
+        assert runtime.calls == []
+
+    asyncio.run(scenario())

docs/README.md

@@ -30,6 +30,7 @@
 
 - [AI Core 与 Agent Core 开发说明](development/AI-Core与Agent-Core开发说明.md)
 - [Knowledge 与 Retrieval Core 开发说明](development/Knowledge与Retrieval-Core开发说明.md)
+- [Benchmark 开发说明](development/Benchmark开发说明.md)
 - [模型提供商与模型发现开发说明](development/模型提供商与模型发现开发说明.md)
 - [MCP Bridge 与 Plugin Host 开发说明](development/MCP-Bridge与Plugin-Host开发说明.md)
 - [独立 MCP Server 配置中心开发说明](development/独立MCP-Server配置中心开发说明.md)

docs/architecture/AI笔记软件技术栈说明-团队版-v2.3.md

@@ -5,7 +5,7 @@
 > 适用范围：桌面客户端、本地知识库、RAG、Agent、Skill、多模型接入、多模态处理与可选云同步  
 > 目标读者：前端、Rust 桌面端、Python AI Core、算法、测试与后续接手项目的开发成员
 
-> 实施状态更新：2026-09-02。本文同时包含目标架构、当前实现和第二阶段接口基线。第一阶段已完成 Vue Web 联调前端、FastAPI、Knowledge/Retrieval、Agent/Tool/Permission、Skill/Plugin 声明式运行时、Mock/OpenAI-Compatible/Ollama Provider、DeepSeek/OpenAI 预设、模型发现及开发阶段 Fernet 凭据存储。Web Workspace 已通过 FastAPI 接入后端配置的真实单 Vault；第二阶段 Agent Trace 持久化、分页快照、可恢复 SSE、stdio MCP Bridge、隔离 Plugin Host、Plugin Command 与 Plugin Settings/Secret Contract 已完成。后续继续接入真实音频处理、Provider 协议增强、Benchmark、文档导出、主题包、Trace 可视化、Mermaid 和函数图像。Tauri/Rust Host、Stronghold、原生多 Vault 文件系统和 Sync Server 仍未实现。
+> 实施状态更新：2026-09-04。本文同时包含目标架构、当前实现和第二阶段接口基线。第一阶段已完成 Vue Web 联调前端、FastAPI、Knowledge/Retrieval、Agent/Tool/Permission、Skill/Plugin 声明式运行时、Mock/OpenAI-Compatible/Ollama Provider、DeepSeek/OpenAI 预设、模型发现及开发阶段 Fernet 凭据存储。Web Workspace 已通过 FastAPI 接入后端配置的真实单 Vault；第二阶段 Agent Trace 持久化、分页快照、可恢复 SSE、stdio MCP Bridge、隔离 Plugin Host、Plugin Command 与 Plugin Settings/Secret Contract 已完成。阶段 E 已完成 Responses/Anthropic 协议、国内 logo 预设、Provider 配置恢复和 Embedding/转写/声纹 API 路由；本地语音模型仍为阶段 F 接口预留。RAG Benchmark 检索评测（Dataset 加载、异步运行、SSE 进度、指标聚合与报告）已完成，Agent Benchmark 暂缓。后续继续接入真实音频处理、文档导出、主题包、Trace 可视化、Mermaid 和函数图像。Tauri/Rust Host、Stronghold、原生多 Vault 文件系统和 Sync Server 仍未实现。
 
 ---
 
@@ -2100,9 +2100,13 @@ MRR
 Citation Hit Rate
 P50 Latency
 P95 Latency
+total_cases
+successful_cases
+failed_cases
+failure_rate
 ```
 
-Benchmark 参数、Embedding 模型、Reranker、数据集版本和运行环境需要一起记录，保证不同实验结果可以复现。
+失败样本按零分计入质量指标分母，报告同时输出样本构成字段标明实际分母。Benchmark 参数、Embedding 模型、Reranker、数据集版本和运行环境需要一起记录，保证不同实验结果可以复现。
 
 ### 20.3 Agent Benchmark
 

docs/contracts/后端接口契约-开发版.md

@@ -180,7 +180,7 @@ RunCancelled
 
 - Chat、Agent Run、Agent Events、Tool 列表、Provider 配置生命周期、模型列表和连接测试已经接入 AI Core。
 - Agent Run/Event 已持久化到 SQLite；SSE 帧携带 sequence `id`，断线后可以回放缺失事件。Trace API 与 Benchmark 共用同一事件事实，并在入库前执行 Secret 脱敏和结果限长。
-- Provider Adapter 当前包含 Mock、真正增量 SSE 的 OpenAI-Compatible Chat Completions，以及 Ollama JSONL Streaming。
+- Provider Adapter 当前包含 Mock、增量 SSE 的 OpenAI-Compatible Chat Completions、OpenAI Responses、Anthropic Messages，以及 Ollama JSONL Streaming。阶段 E 增加 `/api/model-routing`、`/api/models/embeddings`、`/api/media/speaker-matches`；具体请求和阶段边界见第二阶段契约 §8.5。
 - Notes、Search、Index、Skills、Plugins、Tasks 和 Provider 生命周期均已接入业务服务。
 - Workspace 已接入后端配置的真实 Vault；文件树、笔记读写、文件/目录新建、重命名和删除不再使用前端 Mock Fallback。
 - Note Move 保留 `note_id`；Citation 的字符偏移统一使用 UTF-16 code unit，供浏览器编辑器直接定位。
@@ -190,3 +190,9 @@ RunCancelled
 - 接入业务模块时保持当前路径和 Contract，不在 Router 中直接实现数据库、Provider 或 Agent 逻辑。
 
 第二阶段开发保持本文件中已有路径兼容，并按 `第二阶段接口契约-开发版.md` 增加子资源、可选字段和事件。接口完成后先更新 OpenAPI 与本文件，再将第二阶段文档中的状态改为已实现。
+
+
+### 前端真实状态补充（2026-09-04）
+
+- `GET /api/index/status` 额外返回 `total_notes: int` 和 `total_blocks: int`，来自当前 SQLite 索引；未建立内容索引时为 0。
+- `GET /api/permissions/policy` 返回 `Record<string, "allow" | "confirm" | "deny">`，值取自后端当前生效的 PermissionPolicy。此接口只读，不提供全局修改能力，运行时权限确认仍使用既有 Agent permission endpoint。

docs/contracts/第二阶段接口契约-开发版.md

@@ -62,9 +62,9 @@
 | Provider | 现有路径 | `/api/providers/*`、`POST /api/chat` | 扩展 | 补齐协议能力和统一行为 |
 | Retrieval | GET/POST | `/api/index/status`、`/api/index/rebuild` | 扩展 | 暴露 Embedding 兼容状态并安全重建向量 |
 | Benchmark | GET | `/api/benchmarks/datasets` | 计划新增 | 枚举受控 Dataset |
-| Benchmark | POST | `/api/benchmarks/rag/runs` | 计划新增 | 创建 RAG Benchmark |
+| Benchmark | POST | `/api/benchmarks/rag/runs` | 已实现 | 创建 RAG Benchmark |
-| Benchmark | POST | `/api/benchmarks/agent/runs` | 计划新增 | 创建 Agent Benchmark |
+| Benchmark | POST | `/api/benchmarks/agent/runs` | 暂缓 | 创建 Agent Benchmark（依赖 Agent Runtime 完成后交付） |
-| Benchmark | GET | `/api/benchmarks/runs` | 计划新增 | 分页获取 Benchmark Run |
+| Benchmark | GET | `/api/benchmarks/runs` | 已实现 | 分页获取 Benchmark Run |
 | Benchmark | GET/POST | `/api/benchmarks/runs/{run_id}/*` | 计划新增 | 查询、订阅、取消和读取报告 |
 | Export | POST | `/api/exports` | 计划新增 | 创建 HTML/PDF/DOCX 导出任务 |
 | Export | GET | `/api/exports` | 计划新增 | 分页获取导出任务 |
@@ -718,6 +718,8 @@ stdio 命令始终以 executable 与 args 数组通过 `shell=False` 启动；
 
 ## 8. Provider Adapter 扩展
 
+> 阶段 E 实施更新（2026-09-04）：OpenAI Responses、Anthropic Messages、Chat Completions 与 Ollama Adapter 已接入；国内提供商 logo 预设、独立凭据输入、配置恢复、Embedding / 转写 / 声纹 API 路由已实现。真实本地语音模型仍属于阶段 F。实现细节见 [模型提供商与模型发现开发说明](../development/模型提供商与模型发现开发说明.md)。
+
 第二阶段不新增平行 Provider CRUD，继续使用第一阶段接口：
 
 ```text
@@ -734,7 +736,7 @@ POST   /api/chat
 
 ### 8.1 ModelInfo 扩展
 
-`GET /api/providers/{provider_id}/models` 的 item 增加可选字段：
+以下为后续计划的可选字段；阶段 E 的 `GET /api/providers/{provider_id}/models` 实际 item 仍只包含 `model`、`display_name`、`capabilities`：
 
 ```json
 {
@@ -774,6 +776,8 @@ Done
 - 浏览器取消 Fetch 或 SSE 后，服务端必须取消上游 Provider 请求。
 - 不支持 reasoning 的 Provider 不发送伪造 ThinkingDelta。
 
+阶段 E 补充：取消或关闭迭代器直接关闭上游连接并传播取消，不向已断开的客户端继续发送 Done。内部带点号、长名称的工具映射为合法的 64 字符以内名称，响应恢复原命名空间，映射在请求内隔离。实际流中断错误码为 `PROVIDER_STREAM_TRUNCATED`；`PROVIDER_INVALID_RESPONSE` 用于无效结构/参数。上面的 `Done.data.status` 适用于真实 HTTP Adapter；开发 Mock 保留原有测试事件。
+
 ### 8.3 Provider 一致性测试 Contract
 
 每个 Adapter 使用相同 Case 描述：
@@ -811,6 +815,25 @@ MODEL_CONTEXT_LENGTH_EXCEEDED
 
 ---
 
+### 8.5 阶段 E 模型路由接口（已实现）
+
+| 方法 | 路径 | 契约 |
+| --- | --- | --- |
+| GET | `/api/model-routing` | `{config, local_backends}` |
+| PUT | `/api/model-routing` | 提交 ModelRoutingConfig，返回递增版本配置 |
+| POST | `/api/models/embeddings` | `{texts: string[]}` → `{vectors, source, model_id, dimensions, fallback_reason}` |
+| POST | `/api/media/speaker-matches` | `{attachment_id, reference_attachment_id}` → `{score, source, fallback_reason}` |
+
+`ModelRoutingConfig` 包含 `version`、`embedding`、`transcription`、`speaker_matching`。每个能力为 null 或 `{provider_id, model, endpoint, dimensions?}`。dimensions 仅 Embedding 使用，范围 1–16384；endpoint 是选定 Provider 下不带查询的路径，不能传第二个 URL。PUT 不提交 GET 返回的 local_backends；版本冲突返回 409 `MODEL_ROUTING_VERSION_CONFLICT`。删除仍被引用的 Provider 返回 409 `PROVIDER_IN_USE`。
+
+本阶段远程能力仅接受 OpenAI Chat / Compatible HTTP 配置，默认路径分别是 `/embeddings`、`/audio/transcriptions`、`/audio/speaker-matches`。最后一个是本项目自定义 multipart 接口，**不是公共 OpenAI 标准协议**：请求 model、file、reference_file，响应有限 0–1 的 score。转写采用 multipart model、file、可选 language，必须返回非空 text。文件来自受控附件目录，限制 25 MiB。
+
+`TranscriptionJob` 新增可选 `source: api|local|sidecar` 和 `fallback_reason`。保留已有转写 Job 路径；`diarization=true` 返回失败 Job，错误为 `DIARIZATION_NOT_IMPLEMENTED`，不能静默忽略。
+
+无配置时调用本地接口；有配置时 API 优先，网络/鉴权/限流/结果无效时回退。本地 Embedding 当前为 hash 占位；本地 ASR / 声纹后端尚未安装时返回 `LOCAL_MODEL_NOT_INSTALLED`，而非伪成功。远程 Embedding 独立索引并检查完整覆盖，模型变化后需重建；不与本地向量混算。
+
+Provider PATCH 支持 provider_type；普通配置持久化到 SQLite，凭据继续独立加密。预设新增 logo_id、description、capabilities，前端图标随应用打包。
+
 ## 9. RAG / Agent Benchmark
 
 Benchmark Service 同时提供 Python 调用接口和本地 HTTP 接口。CLI、测试和前端报告页调用同一 Service，不各自实现指标。
@@ -889,7 +912,7 @@ Dataset 从仓库或受控导入目录注册。API 不接受调用方提交任
 
 配置快照必须记录 Embedding model ID/version/dimension、Reranker、索引版本、Dataset Hash 和运行环境。
 
-### 9.5 创建 Agent Benchmark
+### 9.5 创建 Agent Benchmark（暂缓，未暴露接口）
 
 `POST /api/benchmarks/agent/runs`
 
@@ -923,12 +946,16 @@ RAG 和 Agent 创建接口均返回 `202 BenchmarkRun`：
   "metrics": null,
   "config_snapshot": {},
   "error": null,
+  "error_code": null,
   "created_at": "2026-08-31T10:30:00Z",
   "started_at": null,
   "completed_at": null
 }
 ```
 
+`status` 取值：`queued` → `running` → `completed` | `failed` | `cancelled`。失败/取消时 `error` 与
+`error_code` 只返回项目错误码与安全消息，不暴露第三方堆栈。
+
 公共接口：
 
 | 方法 | 路径 | 用途 |
@@ -939,6 +966,10 @@ RAG 和 Agent 创建接口均返回 `202 BenchmarkRun`：
 | POST | `/api/benchmarks/runs/{run_id}/cancel` | 取消运行 |
 | GET | `/api/benchmarks/runs/{run_id}/report` | 获取结构化完整报告 |
 
+SSE 事件流（`RunStarted` → `CaseCompleted`* → `RunCompleted` | `RunFailed` | `RunCancelled`）：
+`GET /api/benchmarks/runs/{run_id}/events` 支持 `Last-Event-ID` 与 `?after_sequence=` 游标恢复，
+`RunCompleted` / `RunFailed` / `RunCancelled` 为终止事件，收到后即断流。
+
 ### 9.7 指标 Contract
 
 RAG：
@@ -951,10 +982,17 @@ RAG：
   "mrr": 0.81,
   "citation_hit_rate": 0.89,
   "p50_latency_ms": 24.5,
-  "p95_latency_ms": 67.3
+  "p95_latency_ms": 67.3,
+  "total_cases": 50,
+  "successful_cases": 48,
+  "failed_cases": 2,
+  "failure_rate": 0.04
 }
 ```
 
+失败样本按零分计入质量指标分母，`total_cases` / `successful_cases` / `failed_cases` /
+`failure_rate` 让报告明确实际分母；延迟仅统计成功样本。
+
 Agent：
 
 ```json
@@ -978,8 +1016,10 @@ BENCHMARK_DATASET_NOT_FOUND
 BENCHMARK_DATASET_INVALID
 BENCHMARK_CONFIG_INVALID
 BENCHMARK_INDEX_INCOMPATIBLE
+BENCHMARK_CAPACITY_EXCEEDED
 BENCHMARK_RUN_NOT_FOUND
 BENCHMARK_RUN_FAILED
+BENCHMARK_CASE_EVALUATION_FAILED
 ```
 
 ### 9.9 Retrieval Profile 与索引兼容
@@ -1527,3 +1567,7 @@ frontend/src/
 ```
 
 目录调整应按实际代码规模渐进进行。Router 只做参数接收和错误映射，状态机、第三方 SDK 与文件处理继续放在 Service/Adapter 层。
+
+### Benchmark Embedding 运行归属（阶段 E 集成修复）
+
+`config_snapshot.local_embedding` 仅表示本地基线；`config_snapshot.embedding` 为 `{ "policy": "per_case", "details": "cases[].embedding" }`。报告与 CaseCompleted 事件的逐样本 `embedding` 包含实际 source（api/local/not_used/unavailable）、model_id、dimensions，以及可选 version、fallback_reason、requested_route、route_version、attempted_space。requested_route 仅含提供商引用、模型、相对端点和维度，不包含 API Key 或凭据引用。FTS 不使用 Embedding，标记 not_used；远程失败或索引不完整回退时记录实际本地模型及原因。

docs/development/AI-Core与Agent-Core开发说明.md

@@ -342,7 +342,7 @@ Skill Manifest
 
 前端智能体页面已经完成中文联调：运行状态、Agent Event、内置 Tool、Permission 和常用事件详情字段均通过集中标签映射展示中文；`notes.search` 等技术 ID 继续保留，便于与后端 Trace、日志和接口契约对应。
 
-- 已实现 Mock、OpenAI-Compatible Chat Completions 与 Ollama Adapter；OpenAI Responses 和 Anthropic Messages 尚未实现。
+- 已实现 Mock、OpenAI-Compatible Chat Completions、Ollama、OpenAI Responses 和 Anthropic Messages Adapter；阶段 E 同时完成国内预设、持久化配置和能力模型路由，详见 [模型提供商与模型发现开发说明](模型提供商与模型发现开发说明.md)。
 - Provider 配置暂存内存，后续通过 Repository 接入 SQLite；PATCH 已支持用显式 `null` 清空 base URL、默认模型和凭据引用。
 - Run/Trace 已通过 Repository 接入 SQLite；后续增加按保留策略归档和 Benchmark 引用保护。
 - Permission 已有核心等待/恢复机制，前端确认 UI 已完成联调和中文展示。

docs/development/Benchmark开发说明.md

@@ -0,0 +1,78 @@
+# Benchmark 开发说明
+
+> 所属模块：Knowledge / Retrieval Core（后端，负责人 yxx）。RAG Benchmark 已交付；Agent Benchmark 暂缓，待 Agent Runtime 完成后在同一契约下补齐。
+
+## 定位
+
+Benchmark Service 用受控 Dataset 对检索引擎做可复现评测：创建即返回 queued、后台 asyncio.Task 执行、SSE 实时推送进度、结束后产出结构化报告。CLI、测试与前端报告页复用同一 Service，不各自实现指标。
+
+## 接口
+
+| 方法 | 路径 | 用途 |
+| --- | --- | --- |
+| GET | `/api/benchmarks/datasets?kind=rag` | 枚举受控目录下的 Dataset 元信息 |
+| POST | `/api/benchmarks/rag/runs` | 创建 RAG Benchmark（202） |
+| GET | `/api/benchmarks/runs?kind=&status=&limit=&offset=` | 分页获取运行记录 |
+| GET | `/api/benchmarks/runs/{run_id}` | 状态与指标摘要 |
+| GET | `/api/benchmarks/runs/{run_id}/events` | SSE 进度与 Case 结果 |
+| POST | `/api/benchmarks/runs/{run_id}/cancel` | 取消运行 |
+| GET | `/api/benchmarks/runs/{run_id}/report` | 结构化完整报告 |
+
+Agent Benchmark 的 `/api/benchmarks/agent/runs` 未暴露（暂缓），不在 OpenAPI 注册占位接口。
+
+## Dataset
+
+Dataset 来自 `settings.benchmark_datasets_path`（默认 `backend/data/benchmarks`），API 不接受调用方提交任意路径。按文件名 stem 精确匹配 `{dataset_id}.json`，与请求无关文件的损坏（JSON 语法错误、UTF-8 解码错误、顶层非对象）不会阻断加载；只有目标文件本身损坏才返回 `BENCHMARK_DATASET_INVALID`。
+
+RAG Case 结构：`case_id`、`query`、`expected_note_ids`、`expected_block_ids`、`citation_required`、`tags`。`citation_required=true` 时必须声明 `expected_block_ids`，否则无法计算 Citation Hit Rate。
+
+## 运行生命周期
+
+`queued → running → completed | failed | cancelled`。
+
+- 创建时校验索引兼容性：索引非空、Embedding model/dim 与当前引擎一致、vector/hybrid 时向量索引非空；不满足返回 `BENCHMARK_INDEX_INCOMPATIBLE`（409），避免把环境/索引错误误判为检索质量差。
+- 内存注册表上限 `MAX_RUNS=100`，超限只淘汰终态 run；满容量且全为活动 run 时返回 `BENCHMARK_CAPACITY_EXCEEDED`（429）。
+- 失败/取消只向公开响应暴露项目错误码与安全消息，详细异常进入日志，不通过 HTTP/SSE 返回。
+
+## 指标
+
+RAG 按 (mode, case, repeat) 逐样本计算，再按 mode 聚合：
+
+- 质量：`hit_at_1`、`hit_at_5`、`recall_at_k`、`mrr`、`citation_hit_rate`；
+- 延迟：`p50_latency_ms`、`p95_latency_ms`（仅统计成功样本）；
+- 样本构成：`total_cases`、`successful_cases`、`failed_cases`、`failure_rate`。
+
+失败样本按零分计入质量指标分母，报告据此可知实际分母，避免把执行失败误判为检索质量差。
+
+## 事件与 SSE
+
+事件流：`RunStarted → CaseCompleted* → RunCompleted | RunFailed | RunCancelled`。
+
+`GET /api/benchmarks/runs/{run_id}/events` 支持 `Last-Event-ID` 与 `?after_sequence=` 游标恢复（复用 Agent SSE 的解析逻辑），`RunCompleted` / `RunFailed` / `RunCancelled` 为终止事件，收到后断流。
+
+## 错误码
+
+```text
+BENCHMARK_DATASET_NOT_FOUND
+BENCHMARK_DATASET_INVALID
+BENCHMARK_INDEX_INCOMPATIBLE
+BENCHMARK_CAPACITY_EXCEEDED
+BENCHMARK_RUN_NOT_FOUND
+BENCHMARK_RUN_FAILED
+BENCHMARK_CASE_EVALUATION_FAILED
+```
+
+## 配置快照
+
+报告与运行记录保存 `config_snapshot`：dataset hash/version、modes、retrieval 参数、Reranker、索引元数据、App 版本与环境、Python 版本。`local_embedding` 记录本地基线 model/version/dim；`embedding.policy = per_case` 表示实际来源以逐样本结果为准，不能把本地基线当作本次使用的模型。
+
+每个 `RAGCaseResult.embedding`（同时出现在报告 cases 和 CaseCompleted SSE 中）记录 `source`（api/local/not_used/unavailable）、实际 `model_id` 空间标识、`dimensions`、本地 `version`、`fallback_reason`。远程路由还记录请求时的 `route_version` 和 `requested_route`（provider_id/model/endpoint/dimensions，不含凭据）、成功生成查询向量后的 `attempted_space`。FTS 标记 not_used；调用失败而未完成向量检索时标记 unavailable。API 不可用或远程索引缺失时，实际模型仍记录最终使用的本地基线。配置允许在样本间改变，逐样本记录对应实际调用；汇总指标可能包含多种空间，比较实验时需检查 cases。记录使用任务局部上下文隔离，并发评测不会相互覆盖。
+
+## 测试
+
+```powershell
+cd backend
+uv run pytest -q
+```
+
+`tests/test_benchmark.py` 覆盖数据集注册与校验、指标纯函数、端到端运行、取消、索引兼容、容量与失败样本聚合；`tests/test_retrieval.py` 覆盖 FTS 阈值与分页 total 一致性。

docs/development/Knowledge与Retrieval-Core开发说明.md

@@ -198,7 +198,7 @@ cd backend
 uv run pytest -q
 ```
 
-当前后端完整测试共 136 个用例通过（单元 + 端到端）。测试通过 `tests/conftest.py` 的 autouse fixture 把
+当前后端完整测试共 218 个用例通过（单元 + 端到端）。测试通过 `tests/conftest.py` 的 autouse fixture 把
 数据目录/DB/Vault 重定向到临时目录，不读写真实 `backend/data`，任何本机状态下结果确定。
 
 ## 配置
@@ -232,4 +232,6 @@ rag.search
 - Embedding / Reranker 为轻量实现，后续替换为真实模型（接口不变）。
 - 小语料下 hybrid 检索召回偏宽（向量 Top-K 覆盖全部 block），可加相关性阈值收紧。
 - 重建为同步 + 全量，后续接入增量索引与异步任务队列。
-- 检索 Benchmark 待建立。
+- RAG Benchmark 已建立：`POST /api/benchmarks/rag/runs` 创建即返回 queued、后台 Task 执行，
+  通过 SSE 实时推送进度，报告含逐 Case 结果与 `total_cases` / `successful_cases` / `failed_cases` / `failure_rate`。
+- Agent Benchmark 暂缓，待 Agent Runtime 完成后交付。

docs/development/前端壳子与接口层开发说明.md

@@ -1,6 +1,6 @@
 # 前端壳子与接口层开发说明
 
-> 更新日期：2026-09-02
+> 更新日期：2026-09-04
 > 适用范围：Vue 3 + TypeScript 页面、Workspace、公共 Service、FastAPI 接口适配和 SSE。
 > 文档用途：帮助团队理解当前前端可用能力、模块边界、启动方式和后续页面开发入口。
 
@@ -153,12 +153,7 @@ Service 已适配当前 FastAPI Contract：
 - 识别 `Done`、`RunCompleted`、`RunFailed` 和 `RunCancelled`；
 - 支持 AbortController 主动取消。
 
-Chat Store 已从定时器模拟输出切换为真实 `/api/chat` SSE。默认离线联调配置为：
+Chat Store 使用真实 `/api/chat` SSE。提供商从后端配置加载，前端不展示后端内置测试 Provider，也不预选模拟模型；模型 ID 使用所选提供商保存的默认值，并支持手动输入。
-
-```text
-provider_id = mock
-model       = mock-1
-```
 
 ## 8. 环境和启动
 
@@ -207,3 +202,34 @@ Vite 当前会提示 Chat 与 Workspace 的部分异步 Chunk 超过 500 kB，
 - Workspace 接入 Tauri 后，需要增加路径规范化、写入失败恢复和外部修改冲突测试；
 - 页面新增交互必须经过键盘、空状态、加载状态、错误状态和窄窗口检查；
 - Workspace 的 Milkdown 写作模式与 CodeMirror 源码模式共享同一 Markdown 数据源；后续修改编辑器时不得改变 Store/Service 边界，并必须保留文件切换、自动保存和选区格式化回归测试。
+
+
+## 阶段 F 前：前端真实数据清理
+
+已删除运行时的聊天示例、Agent Run/Event/Tool/权限示例、Provider/Model、Task、Skill、Plugin、IndexStatus 常量和 searchMock。测试文件中的隔离桩保留，仅用于自动化验证。
+
+- 所有业务 Store 从空集合开始，由真实 API 填充；连接失败显示错误，不回退演示记录。
+- 普通聊天仅显示用户实际输入和 SSE 响应；当前会话列表保留在页面会话内，刷新后清空，后端暂无聊天历史持久化接口。切换会话保留本次会话内的真实消息，取消旧流并屏蔽迟到回调。
+- 聊天页移除尚未接入的知识库与 Skill 开关，知识库工具和 Skill 通过 Agent 使用。
+- 设置页不再伪造健康状态、版本、42 篇笔记/318 个 Block、模型名称和索引能力开关。状态未获取时显示 unknown/未获取；应用版本来自 package.json，后端版本来自 /api/status。
+- GET /api/index/status 增加 total_notes、total_blocks，直接读取 SQLite 的当前索引统计。
+- GET /api/permissions/policy 返回 PermissionPolicy 的实际生效值。设置页只读展示；全局策略编辑暂未开放，运行权限确认仍走原有 Agent 接口。
+- 删除模拟重启成功逻辑，说明 Web 端不具备进程重启能力；索引页面只保留后端已实现的全量重建。
+- Task DTO 不再填充后端未返回的优先级和来源，Agent Token 用量不再把未知输入/输出拆分填成 0。
+- Plugin/Skill/Provider 无记录时显示空状态，模型发现失败时允许使用真实的手动模型 ID。
+
+验证：前端 81 项测试、类型检查与生产构建通过；后端 454 项测试通过。新增测试覆盖空初始状态、离线错误、真实统计与权限、测试 Provider 过滤、真实聊天历史及旧流隔离。本次未调用真实付费推理 API。
+
+### MCP 工具中文展示补充
+
+Agent 工具列表按 `mcp.<server_id>.<remote_name>` 的远程工具名匹配中文展示，支持 `web_search`（网页搜索）、`understand_image`（图像理解），并补充 `text.uppercase`（文本转大写）。此映射只影响界面，工具调用与权限选择仍使用完整原始 ID。
+
+卡片默认显示三行摘要，完整服务原文可展开查看，展开操作不会改变工具选择。服务已提供中文说明时优先保留；未收录的 MCP 工具明确提示暂无中文说明，不将本地摘要当作服务协议或自动翻译结果。原始说明及其中的参数规则完整保留。
+
+验证：前端 84 项测试、类型检查与生产构建通过。新增回归覆盖不同服务器命名空间、未知工具、服务中文说明、原文完整性，以及选择工具时保留原始 ID。
+
+### 聊天模型选择审阅修复
+
+返回聊天页时保留仍启用的提供商与手动模型 ID，仅刷新其模型列表；未选择、已删除或已禁用的提供商才回退到默认值。提供商加载失败时保留当前选择并展示错误。新增页面重新挂载与异常分支回归，前端共 89 项测试通过。
+
+补充卸载时序修复：提供商或技能加载期间离开聊天页后，旧页面的初始化回调不再修改聊天选择，迟到错误也不再更新旧页面。两种加载延迟均通过先失败、修复后通过的回归测试，并验证返回页面后的默认模型和发送按钮状态；前端共 91 项测试通过。

docs/development/模型提供商与模型发现开发说明.md

@@ -1,107 +1,107 @@
-# 模型提供商与模型发现开发说明
+# 模型提供商、协议适配与模型路由开发说明
 
-> 更新日期：2026-09-02。OpenAI、DeepSeek、Ollama 预设、模型自动发现、默认模型选择和开发阶段加密凭据存储均已实现并接入设置页。
+> 更新日期：2026-09-04。阶段 E 实现记录。本地小模型的实际安装与多模态队列属于阶段 F；本阶段保留并测试可注入的本地后端接口。
 
-## 1. 本次目标
+## 1. 设置与凭据
 
-本次完善设置页的模型提供商配置，不改变 Agent、Chat 和 Skill 对统一 Model Core 接口的依赖：
+设置 → 模型提供商 → 新增 Provider 提供可搜索的 logo 预设网格，包含 DeepSeek、Kimi、阿里云百炼、智谱 GLM、火山方舟、硅基流动、百度千帆、腾讯混元、MiniMax、阶跃星辰，以及 OpenAI Chat / Responses、Anthropic 和 Ollama。图标打包到前端，使用时不请求第三方图片服务；来源和许可见前端 assets/providers 目录。
 
-- 提供 OpenAI、DeepSeek 和 Ollama 配置预设；
+预设返回 `preset_id`、`logo_id`、`name`、`provider_type`、`base_url`、`requires_credential`、`description` 和 `capabilities`。能力标签表示预设接入范围，不保证该账号的每个模型支持全部能力。厂商专用媒体协议、Coding Plan 和海外地域需要使用对应地址，不能仅凭厂商名称推断协议兼容。
-- 保存 Provider 后自动获取该账号或服务当前可用的模型列表；
-- 支持手动刷新模型列表和选择默认模型；
-- 保留自定义 OpenAI-Compatible 服务入口；
-- 不在 Vue、FastAPI 配置或仓库文件中保存、回显 API Key 明文。
 
-## 2. 接口与实现
+预设和自定义服务都可以直接输入 API Key。每个新配置分配独立 Credential ID，避免同厂商多账号相互覆盖。明文只留在密码输入框和专用请求中，提交、失败、切换预设及关闭时清空；密钥不进入 Pinia、localStorage、Provider 配置响应或模型路由。
 
-### 2.1 Provider 预设
+凭据继续使用独立的 `PUT /api/credentials/{credential_id}` 和 Fernet 开发存储。`plugin.*`、`mcp.*` 是保留命名空间。桌面端阶段仍需要把主密钥管理迁移到 Stronghold。保存密钥与保存 Provider 是两个请求，Provider 保存失败时可能留下未引用的加密凭据，可通过凭据删除接口清理。
 
-新增接口：
+Provider 配置和 Credential ID 写入 SQLite `provider_configs`，重启后恢复。Mock 为内置 Provider，不能编辑或删除。PATCH 已支持变更 `provider_type` 并重新创建 Adapter；Base URL 限制为不带用户信息、查询或 fragment 的 HTTP(S) 地址。
 
-```http
+## 2. 协议适配
-GET /api/providers/presets
+
+支持的协议是 OpenAI Chat Completions、OpenAI-Compatible、OpenAI Responses、Anthropic Messages 和 Ollama。Agent、Chat、Skill 仍只依赖内部 `ModelRequest` / `ModelEvent` / `ProviderTurn`，不直接解释厂商协议。
+
+Adapter 负责消息及 Tool 历史转换、增量文本、可用的 reasoning delta、工具参数片段、usage、终止与统一错误。外部错误正文不原样返回；HTTP 鉴权、限流、超时、无效数据、流中断分别映射为内部错误。取消继续传播并关闭上游连接，不触发第二次本地推理。
+
+`GET /api/providers/{provider_id}/models` 用于发现模型。模型列表不等于每个模型的能力承诺；部分厂商或代理不提供 `/models` 时，允许直接手动输入模型 ID。连接测试验证模型发现接口，不代表每一种媒体模型已完成真实推理验收。
+
+## 3. 三类模型路由
+
+接口：
+
+| 方法 | 路径 | 用途 |
+| --- | --- | --- |
+| GET | `/api/model-routing` | 读取配置和本地后端状态 |
+| PUT | `/api/model-routing` | 带版本更新三类模型绑定 |
+| POST | `/api/models/embeddings` | 文本向量，返回来源和回退原因 |
+| POST | `/api/media/transcriptions` | 附件转写作业 |
+| GET | `/api/media/transcriptions/{job_id}` | 获取转写作业 |
+| POST | `/api/media/speaker-matches` | 两个音频附件的声纹相似度 |
+
+设置 → 索引与模型分别选择 Embedding、音频转文本和声纹匹配。三种绑定互相独立，可使用不同提供商、模型、密钥和 API 路径。
+
+GET / PUT 响应：
+
+```json
+{
+  "config": {
+    "version": 1,
+    "embedding": {
+      "provider_id": "provider_example",
+      "model": "your-embedding-model",
+      "endpoint": "/embeddings",
+      "dimensions": null
+    },
+    "transcription": null,
+    "speaker_matching": null
+  },
+  "local_backends": [
+    {"capability": "embedding", "status": "placeholder", "message": "当前为 hash-v1 占位向量"},
+    {"capability": "transcription", "status": "not_installed", "message": "阶段 F 接入"},
+    {"capability": "speaker_matching", "status": "not_installed", "message": "阶段 F 接入"}
+  ]
+}
 ```
 
-预设由后端 `ProviderFactory` 提供，前端只消费名称、协议类型、Base URL 和是否需要凭据等配置元数据，不直接实现厂商协议。
+PUT body 只提交 `config` 的内容。`version` 为读取时的版本，成功递增；并发更新返回 `MODEL_ROUTING_VERSION_CONFLICT`。绑定为空表示使用本地后端。删除仍被路由引用的 Provider 返回 `PROVIDER_IN_USE`，须先解除绑定。
 
-当前预设：
+本阶段三类远程路由使用 `openai_chat` / `openai_compatible` 的 Bearer HTTP 配置，endpoint 只能是该提供商下的路径。Responses、Anthropic 和 Ollama 原生协议不冒充上述媒体协议；Ollama 用户需要另建兼容 HTTP 配置才能用于当前远程 Embedding 接口。
 
-| 提供商 | Provider Type | Base URL | 默认 Credential ID |
+调用规则：无绑定 → 本地接口；有绑定 → API → 校验结果 → 失败或无效时调用本地接口。Provider 停用、密钥缺失、鉴权失败、限流、网络超时及无效结果均可回退；用户取消不会回退。附件不存在、大小非法等输入错误直接返回，不把用户输入错误当成模型故障。
-| --- | --- | --- | --- |
-| OpenAI | `openai_chat` | `https://api.openai.com/v1` | `openai` |
-| DeepSeek | `openai_compatible` | `https://api.deepseek.com` | `deepseek` |
-| Ollama | `ollama` | `http://127.0.0.1:11434` | 无 |
 
-OpenAI 和 DeepSeek 都通过项目已有的 `OpenAICompatibleProvider` 访问。模型发现分别请求 Base URL 下的 `/models`，不引入厂商 SDK。
+## 4. Embedding 与索引一致性
 
-### 2.2 自动获取模型
+请求使用 `model`、`input`、`encoding_format: float`；只有明确配置维度时才发送 `dimensions`。按最多 32 条分批请求，全部批次有效才使用 API 结果。校验返回数量、连续唯一 index、维度一致性、有限数值、非零范数，并 L2 归一化。维度可为 1–16384，不截断、补零或混用不同模型的向量。
 
-模型列表继续使用既有接口：
+返回 `vectors`、`source`、`model_id`、`dimensions`、`fallback_reason`。远程空间 ID 由完整 API URL、模型和实际维度生成；即使维度相同，不同模型的空间也不同。
 
-```http
+笔记索引始终保留现有 hash/sqlite-vec 本地基线，远程向量写入独立 `routed_block_vectors` 表。远程查询只搜索对应空间，并要求覆盖全部当前 Block。API 失败、索引缺失、不完整或损坏时使用完整本地索引。切换模型、URL、维度后应在设置中重建全部索引。旧空间与当前文本不会混合打分，删除笔记或重建索引会通过外键清理远程向量。
-GET /api/providers/{provider_id}/models
-```
 
-设置页在以下时机调用该接口：
+当前远程侧索引采用 SQLite JSON 向量和精确余弦扫描，复杂度 O(Block 数量 × 维度)，适用于当前小型 Vault；后续大规模索引需替换为按空间隔离的 ANN。网络等待发生在数据库写事务之前，当前仍会增加保存或重建延迟，异步索引队列尚未接入。全量重建先在内存中准备全部向量，再使用一个 SQLite 事务更新元数据、FTS、本地与远程向量及任务关联；取消或失败只回滚索引事务，不再覆盖整库文件。准备阶段保留旧索引可查询，代价是内存同时容纳本次重建的向量。
 
-- Provider 列表加载完成后，为所有已启用 Provider 自动刷新；
+OpenAI Compatible 流中，工具名称可能分片返回。适配器在本轮输出结束后发送完整工具名及已缓冲参数，避免把名称片段当作工具 ID；文本与推理内容仍逐片发送。
-- 新增或编辑 Provider 保存成功后自动刷新；
-- 用户点击“刷新模型”时手动刷新；
-- 打开已有 Provider 的编辑窗口时刷新可选模型。
 
-前端按模型名称排序并按 `model_id` 去重。获取结果保存在 `providerStore.modelsByProvider`，加载状态和错误按 Provider 隔离，单个外部服务失败不会阻止其他服务展示。
+无 API 时使用的 `HashEmbeddingProvider` 是确定性特征哈希占位实现，**不是已集成的小型语义模型**。真实本地 Embedding 可实现既有 `EmbeddingProvider` 接口注入。
 
-获取成功后，Provider 卡片展示模型数量和默认模型下拉框。更换默认模型会调用 Provider PATCH 接口写回配置；编辑窗口仍允许手动输入模型 ID，以兼容未出现在列表中的代理模型或部署别名。
+## 5. 音频与声纹边界
 
-### 2.3 错误处理
+转写默认请求 `/audio/transcriptions`，multipart 字段 `model`、可选 `language` 和 `file`，响应必须包含非空字符串 `text`。已有纯文本附件和 Host 旁路 `.txt` 导入保留，来源标记 `sidecar`，不伪称 ASR。转写作业新增 `source`、`fallback_reason`；回退失败的作业记录 `LOCAL_MODEL_NOT_INSTALLED` 等明确错误。作业目前同步执行、限量保存在内存中，不是持久化异步队列。
 
-Provider Adapter 的错误在 FastAPI 路由转换为统一 API Error：
+声纹匹配使用**本项目自定义 HTTP 契约**，默认 `/audio/speaker-matches`，multipart 字段 `model`、`file`、`reference_file`；响应为 `{"score": 0.85}`，score 必须为有限的 0–1 数值。公共入口只接受 `attachment_id` 和 `reference_attachment_id`，不接收任意文件路径。此接口用于一对一声纹比对，不等同于 pyannote 说话人分离，也不声称任意国内厂商原生支持该路径。
 
-| Provider Error | HTTP 状态 |
+媒体文件限制 1 字节至 25 MiB，API 响应限制 16 MiB，单次请求超时 30 秒。文件从后端受控附件目录读取，使用结束或取消时关闭句柄。
-| --- | --- |
-| `PROVIDER_AUTH_FAILED` | 401 |
-| `MODEL_NOT_FOUND` | 404 |
-| `PROVIDER_RATE_LIMITED` | 429 |
-| `PROVIDER_TIMEOUT` | 504 |
-| 其他 Provider 可用性错误 | 502 |
 
-前端在对应 Provider 卡片内展示失败原因，并允许用户修正 Credential ID、Base URL 后重新获取。
+`LocalSpeechBackend` 提供 `transcribe` 和 `match` 接口。阶段 E 默认 `PendingSpeechBackend` 明确报告未安装；阶段 F 接入 faster-whisper、pyannote.audio 及模型资源后替换。当前 `diarization=true` 明确返回失败作业 `DIARIZATION_NOT_IMPLEMENTED`，不会静默忽略。视频解码、TTS、视频生成及厂商专用异步媒体协议不在本次交付内。
 
-## 3. 凭据边界
+## 6. 官方协议依据与验证
 
-设置页选择 OpenAI 或 DeepSeek 预设后展示密码类型的 API Key 输入框，不再要求用户理解 Credential ID。输入值只存在于表单的临时 `ref`，不会写入 Pinia 或 localStorage；请求完成、取消表单或失败后都会清空。
+国内通用地址核对依据：[阿里云百炼兼容接口](https://help.aliyun.com/zh/model-studio/compatibility-of-openai-with-dashscope)、[百度千帆兼容接口](https://cloud.baidu.com/doc/qianfan/s/Hmh4suq26)、[腾讯混元兼容接口](https://cloud.tencent.com/document/product/1729/111007)、[MiniMax 文本接口](https://platform.minimaxi.com/docs/guides/text-generation)、[阶跃星辰通用与套餐地址区别](https://platform.stepfun.com/docs/zh/step-plan/overview)、[火山方舟 API](https://www.volcengine.com/docs/82379/1795150)、[智谱开放接口](https://docs.bigmodel.cn/api-reference/文件-api/文件列表)。模型 ID 以账号实际开通列表为准，不写死“最新模型”。
 
-API Key 通过独立接口写入：
+流式事件依据：[OpenAI Responses streaming](https://platform.openai.com/docs/api-reference/responses-streaming)、[Anthropic streaming](https://platform.claude.com/docs/en/build-with-claude/streaming)。音频请求依据：[SiliconFlow transcription](https://docs.siliconflow.com/en/api-reference/audio/create-audio-transcriptions)。
 
-```http
+自动化验证使用虚构凭据、本地附件、httpx.MockTransport 和可注入本地模型，覆盖流式 Tool/Usage/取消、错误映射、回退、索引空间隔离、版本冲突、重启恢复和界面凭据行为。没有使用真实 API Key 或向厂商发送推理请求。审阅修复并同步主分支后验证：后端全量 447 项、前端 76 项测试通过，Vue/TypeScript 类型检查和生产构建通过，浅色/深色预设页面与路由保存经过浏览器检查，git diff --check 通过。后端仅保留既有 Starlette 测试客户端弃用提示，前端保留既有大 bundle 提示。
-GET    /api/credentials/{credential_id}
-PUT    /api/credentials/{credential_id}
-DELETE /api/credentials/{credential_id}
-```
 
-PUT 请求使用 Pydantic `SecretStr` 接收密钥，响应仅包含 Credential ID 和 `configured` 状态。后端使用 Fernet 认证加密，将密文保存到 `data/credentials/credentials.json`，主密钥保存到 `data/credentials/master.key`；目录和文件尽可能设置为仅当前用户可访问并整体排除版本控制。写入采用临时文件替换，避免进程中断留下半写文件。Provider 发起请求时按 Credential ID 解密，解密失败转换为统一 Provider Error，任何读取接口均不返回明文。
+```powershell
-
-本地开发存储的主密钥与密文仍位于同一用户数据目录，因此它解决的是仓库泄漏、普通配置误提交和静态明文暴露，不等同于操作系统安全硬件或 Stronghold。Tauri 集成后应以 Stronghold 实现替换 `EncryptedCredentialStore`。无界面环境仍兼容 `OPENAI_API_KEY`、`DEEPSEEK_API_KEY` 和 Host 注入的 `AINOTE_CREDENTIAL_<ID>`；设置页保存的本地密钥优先，环境变量仅作为回退。
-
-自动化测试仅使用虚构测试值，验证磁盘文件不包含明文、加解密往返、API 响应不泄密，以及 Provider 能用解密后的值构造 Authorization Header。本次没有使用真实 OpenAI 或 DeepSeek Key，也没有向厂商发起真实请求。
-
-## 4. 验证
-
-后端：
-
-```bash
 cd backend
 uv run pytest -q -p no:cacheprovider
-```
+cd ../frontend
-
-前端：
-
-```bash
-cd frontend
 pnpm test
 pnpm build
 ```
-
-自动化验证覆盖 Provider 预设、OpenAI-Compatible `/models` 请求与鉴权头、模型映射、前端自动刷新、排序去重及按 Provider 隔离错误。生产构建同时执行 Vue 和 TypeScript 类型检查。
-
-当前完整回归基线：后端 136 项测试、前端 29 项测试通过，前端类型检查和生产构建通过。Provider 配置目前仍保存在内存 Registry，AI Core 重启后需要重新创建；凭据密文会保留。`plugin.*` 为 Plugin Secret 保留命名空间，Provider 配置、临时测试凭据和通用凭据 API 均拒绝该前缀。OpenAI Responses 与 Anthropic Messages Adapter 尚未实现，设置页正式预设不会使用这两种协议。

frontend/pnpm-workspace.yaml

@@ -0,0 +1,2 @@
+allowBuilds:
+  esbuild: true

frontend/public/provider-icons-LICENSE.txt

@@ -0,0 +1,40 @@
+Lobe Icons — Copyright (c) 2023 LobeHub. MIT license; see LICENSE.
+Source: https://github.com/lobehub/lobe-icons
+Revision: 4aaf4ee1fb2678a7f989ea570f0f6ce14a9abf75
+Source directory: packages/static-svg/icons/
+Assets are bundled locally. Brand trademarks belong to their respective owners.
+File mapping (local: upstream):
+anthropic.svg: anthropic.svg
+baidu.svg: baidu-color.svg
+deepseek.svg: deepseek-color.svg
+hunyuan.svg: hunyuan-color.svg
+kimi.svg: kimi-color.svg
+minimax.svg: minimax-color.svg
+ollama.svg: ollama.svg
+openai.svg: openai.svg
+qwen.svg: qwen-color.svg
+siliconflow.svg: siliconcloud-color.svg
+stepfun.svg: stepfun-color.svg
+volcengine.svg: volcengine-color.svg
+zhipu.svg: zhipu-color.svg
+MIT License
+
+Copyright (c) 2023 LobeHub
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

frontend/src/assets/providers/ATTRIBUTION.txt

@@ -0,0 +1,19 @@
+Lobe Icons — Copyright (c) 2023 LobeHub. MIT license; see LICENSE.
+Source: https://github.com/lobehub/lobe-icons
+Revision: 4aaf4ee1fb2678a7f989ea570f0f6ce14a9abf75
+Source directory: packages/static-svg/icons/
+Assets are bundled locally. Brand trademarks belong to their respective owners.
+File mapping (local: upstream):
+anthropic.svg: anthropic.svg
+baidu.svg: baidu-color.svg
+deepseek.svg: deepseek-color.svg
+hunyuan.svg: hunyuan-color.svg
+kimi.svg: kimi-color.svg
+minimax.svg: minimax-color.svg
+ollama.svg: ollama.svg
+openai.svg: openai.svg
+qwen.svg: qwen-color.svg
+siliconflow.svg: siliconcloud-color.svg
+stepfun.svg: stepfun-color.svg
+volcengine.svg: volcengine-color.svg
+zhipu.svg: zhipu-color.svg

frontend/src/assets/providers/LICENSE

@@ -0,0 +1,21 @@
+MIT License
+
+Copyright (c) 2023 LobeHub
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in all
+copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+SOFTWARE.

frontend/src/assets/providers/anthropic.svg

@@ -0,0 +1 @@
+<svg fill="currentColor" fill-rule="evenodd" height="1em" style="flex:none;line-height:1" viewBox="0 0 24 24" width="1em" xmlns="http://www.w3.org/2000/svg"><title>Anthropic</title><path d="M13.827 3.52h3.603L24 20h-3.603l-6.57-16.48zm-7.258 0h3.767L16.906 20h-3.674l-1.343-3.461H5.017l-1.344 3.46H0L6.57 3.522zm4.132 9.959L8.453 7.687 6.205 13.48H10.7z"></path></svg>

frontend/src/assets/providers/baidu.svg

@@ -0,0 +1 @@
+<svg height="1em" style="flex:none;line-height:1" viewBox="0 0 24 24" width="1em" xmlns="http://www.w3.org/2000/svg"><title>Baidu</title><path d="M8.859 11.735c1.017-1.71 4.059-3.083 6.202.286 1.579 2.284 4.284 4.397 4.284 4.397s2.027 1.601.73 4.684c-1.24 2.956-5.64 1.607-6.005 1.49l-.024-.009s-1.746-.568-3.776-.112c-2.026.458-3.773.286-3.773.286l-.045-.001c-.328-.01-2.38-.187-3.001-2.968-.675-3.028 2.365-4.687 2.592-4.968.226-.288 1.802-1.37 2.816-3.085zm.986 1.738v2.032h-1.64s-1.64.138-2.213 2.014c-.2 1.252.177 1.99.242 2.148.067.157.596 1.073 1.927 1.342h3.078v-7.514l-1.394-.022zm3.588 2.191l-1.44.024v3.956s.064.985 1.44 1.344h3.541v-5.3h-1.528v3.979h-1.46s-.466-.068-.553-.447v-3.556zM9.82 16.715v3.06H8.58s-.863-.045-1.126-1.049c-.136-.445.02-.959.088-1.16.063-.203.353-.671.951-.85H9.82zm9.525-9.036c2.086 0 2.646 2.06 2.646 2.742 0 .688.284 3.597-2.309 3.655-2.595.057-2.704-1.77-2.704-3.08 0-1.374.277-3.317 2.367-3.317zM4.24 6.08c1.523-.135 2.645 1.55 2.762 2.513.07.625.393 3.486-1.975 4-2.364.515-3.244-2.249-2.984-3.544 0 0 .28-2.797 2.197-2.969zm8.847-1.483c.14-1.31 1.69-3.316 2.931-3.028 1.236.285 2.367 1.944 2.137 3.37-.224 1.428-1.345 3.313-3.095 3.082-1.748-.226-2.143-1.823-1.973-3.424zM9.425 1c1.307 0 2.364 1.519 2.364 3.398 0 1.879-1.057 3.4-2.364 3.4s-2.367-1.521-2.367-3.4C7.058 2.518 8.118 1 9.425 1z" fill="#2932E1" fill-rule="nonzero"></path></svg>

frontend/src/assets/providers/deepseek.svg

@@ -0,0 +1 @@
+<svg height="1em" style="flex:none;line-height:1" viewBox="0 0 24 24" width="1em" xmlns="http://www.w3.org/2000/svg"><title>DeepSeek</title><path d="M23.748 4.482c-.254-.124-.364.113-.512.234-.051.039-.094.09-.137.136-.372.397-.806.657-1.373.626-.829-.046-1.537.214-2.163.848-.133-.782-.575-1.248-1.247-1.548-.352-.156-.708-.311-.955-.65-.172-.241-.219-.51-.305-.774-.055-.16-.11-.323-.293-.35-.2-.031-.278.136-.356.276-.313.572-.434 1.202-.422 1.84.027 1.436.633 2.58 1.838 3.393.137.093.172.187.129.323-.082.28-.18.552-.266.833-.055.179-.137.217-.329.14a5.526 5.526 0 01-1.736-1.18c-.857-.828-1.631-1.742-2.597-2.458a11.365 11.365 0 00-.689-.471c-.985-.957.13-1.743.388-1.836.27-.098.093-.432-.779-.428-.872.004-1.67.295-2.687.684a3.055 3.055 0 01-.465.137 9.597 9.597 0 00-2.883-.102c-1.885.21-3.39 1.102-4.497 2.623C.082 8.606-.231 10.684.152 12.85c.403 2.284 1.569 4.175 3.36 5.653 1.858 1.533 3.997 2.284 6.438 2.14 1.482-.085 3.133-.284 4.994-1.86.47.234.962.327 1.78.397.63.059 1.236-.03 1.705-.128.735-.156.684-.837.419-.961-2.155-1.004-1.682-.595-2.113-.926 1.096-1.296 2.746-2.642 3.392-7.003.05-.347.007-.565 0-.845-.004-.17.035-.237.23-.256a4.173 4.173 0 001.545-.475c1.396-.763 1.96-2.015 2.093-3.517.02-.23-.004-.467-.247-.588zM11.581 18c-2.089-1.642-3.102-2.183-3.52-2.16-.392.024-.321.471-.235.763.09.288.207.486.371.739.114.167.192.416-.113.603-.673.416-1.842-.14-1.897-.167-1.361-.802-2.5-1.86-3.301-3.307-.774-1.393-1.224-2.887-1.298-4.482-.02-.386.093-.522.477-.592a4.696 4.696 0 011.529-.039c2.132.312 3.946 1.265 5.468 2.774.868.86 1.525 1.887 2.202 2.891.72 1.066 1.494 2.082 2.48 2.914.348.292.625.514.891.677-.802.09-2.14.11-3.054-.614zm1-6.44a.306.306 0 01.415-.287.302.302 0 01.2.288.306.306 0 01-.31.307.303.303 0 01-.304-.308zm3.11 1.596c-.2.081-.399.151-.59.16a1.245 1.245 0 01-.798-.254c-.274-.23-.47-.358-.552-.758a1.73 1.73 0 01.016-.588c.07-.327-.008-.537-.239-.727-.187-.156-.426-.199-.688-.199a.559.559 0 01-.254-.078c-.11-.054-.2-.19-.114-.358.028-.054.16-.186.192-.21.356-.202.767-.136 1.146.016.352.144.618.408 1.001.782.391.451.462.576.685.914.176.265.336.537.445.848.067.195-.019.354-.25.452z" fill="#4D6BFE"></path></svg>

frontend/src/assets/providers/hunyuan.svg

@@ -0,0 +1 @@
+<svg height="1em" style="flex:none;line-height:1" viewBox="0 0 24 24" width="1em" xmlns="http://www.w3.org/2000/svg"><title>Hunyuan</title><circle cx="12" cy="12" fill="#0055E9" r="12"></circle><path d="M12 0c.518 0 1.028.033 1.528.096A6.188 6.188 0 0112.12 12.28l-.12.001c-2.99 0-5.242 2.179-5.554 5.11-.223 2.086.353 4.412 2.242 6.146C3.672 22.1 0 17.479 0 12 0 5.373 5.373 0 12 0z" fill="#A8DFF5"></path><path d="M5.286 5a2.438 2.438 0 01.682 3.38c-3.962 5.966-3.215 10.743 2.648 15.136C3.636 22.056 0 17.452 0 12c0-1.787.39-3.482 1.09-5.006.253-.435.525-.872.817-1.311A2.438 2.438 0 015.286 5z" fill="#0055E9"></path><path d="M12.98.04c.272.021.543.053.81.093.583.106 1.117.254 1.538.44 6.638 2.927 8.07 10.052 1.748 15.642a4.125 4.125 0 01-5.822-.358c-1.51-1.706-1.3-4.184.357-5.822.858-.848 3.108-1.223 4.045-2.441 1.257-1.634 2.122-6.009-2.523-7.506L12.98.039z" fill="#00BCFF"></path><path d="M13.528.096A6.187 6.187 0 0112 12.281a5.75 5.75 0 00-1.71.255c.147-.905.595-1.784 1.321-2.501.858-.848 3.108-1.223 4.045-2.441 1.27-1.651 2.14-6.104-2.676-7.554.184.014.367.033.548.056z" fill="#ECECEE"></path></svg>

frontend/src/assets/providers/kimi.svg

@@ -0,0 +1 @@
+<svg height="1em" style="flex:none;line-height:1" viewBox="0 0 24 24" width="1em" xmlns="http://www.w3.org/2000/svg"><title>Kimi</title><path d="M21.846 0a1.923 1.923 0 110 3.846H20.15a.226.226 0 01-.227-.226V1.923C19.923.861 20.784 0 21.846 0z" fill="#1783FF"></path><path d="M11.065 11.199l7.257-7.2c.137-.136.06-.41-.116-.41H14.3a.164.164 0 00-.117.051l-7.82 7.756c-.122.12-.302.013-.302-.179V3.82c0-.127-.083-.23-.185-.23H3.186c-.103 0-.186.103-.186.23V19.77c0 .128.083.23.186.23h2.69c.103 0 .186-.102.186-.23v-3.25c0-.069.025-.135.069-.178l2.424-2.406a.158.158 0 01.205-.023l6.484 4.772a7.677 7.677 0 003.453 1.283c.108.012.2-.095.2-.23v-3.06c0-.117-.07-.212-.164-.227a5.028 5.028 0 01-2.027-.807l-5.613-4.064c-.117-.078-.132-.279-.028-.381z" fill="#fff"></path></svg>

frontend/src/assets/providers/minimax.svg

@@ -0,0 +1 @@
+<svg height="1em" style="flex:none;line-height:1" viewBox="0 0 24 24" width="1em" xmlns="http://www.w3.org/2000/svg"><title>Minimax</title><defs><linearGradient id="lobe-icons-minimax-_R_0_" x1="0%" x2="100.182%" y1="50.057%" y2="50.057%"><stop offset="0%" stop-color="#E2167E"></stop><stop offset="100%" stop-color="#FE603C"></stop></linearGradient></defs><path d="M16.278 2c1.156 0 2.093.927 2.093 2.07v12.501a.74.74 0 00.744.709.74.74 0 00.743-.709V9.099a2.06 2.06 0 012.071-2.049A2.06 2.06 0 0124 9.1v6.561a.649.649 0 01-.652.645.649.649 0 01-.653-.645V9.1a.762.762 0 00-.766-.758.762.762 0 00-.766.758v7.472a2.037 2.037 0 01-2.048 2.026 2.037 2.037 0 01-2.048-2.026v-12.5a.785.785 0 00-.788-.753.785.785 0 00-.789.752l-.001 15.904A2.037 2.037 0 0113.441 22a2.037 2.037 0 01-2.048-2.026V18.04c0-.356.292-.645.652-.645.36 0 .652.289.652.645v1.934c0 .263.142.506.372.638.23.131.514.131.744 0a.734.734 0 00.372-.638V4.07c0-1.143.937-2.07 2.093-2.07zm-5.674 0c1.156 0 2.093.927 2.093 2.07v11.523a.648.648 0 01-.652.645.648.648 0 01-.652-.645V4.07a.785.785 0 00-.789-.78.785.785 0 00-.789.78v14.013a2.06 2.06 0 01-2.07 2.048 2.06 2.06 0 01-2.071-2.048V9.1a.762.762 0 00-.766-.758.762.762 0 00-.766.758v3.8a2.06 2.06 0 01-2.071 2.049A2.06 2.06 0 010 12.9v-1.378c0-.357.292-.646.652-.646.36 0 .653.29.653.646V12.9c0 .418.343.757.766.757s.766-.339.766-.757V9.099a2.06 2.06 0 012.07-2.048 2.06 2.06 0 012.071 2.048v8.984c0 .419.343.758.767.758.423 0 .766-.339.766-.758V4.07c0-1.143.937-2.07 2.093-2.07z" fill="url(#lobe-icons-minimax-_R_0_)" fill-rule="nonzero"></path></svg>

frontend/src/assets/providers/ollama.svg

@@ -0,0 +1 @@
+<svg fill="currentColor" fill-rule="evenodd" height="1em" style="flex:none;line-height:1" viewBox="0 0 24 24" width="1em" xmlns="http://www.w3.org/2000/svg"><title>Ollama</title><path d="M7.905 1.09c.216.085.411.225.588.41.295.306.544.744.734 1.263.191.522.315 1.1.362 1.68a5.054 5.054 0 012.049-.636l.051-.004c.87-.07 1.73.087 2.48.474.101.053.2.11.297.17.05-.569.172-1.134.36-1.644.19-.52.439-.957.733-1.264a1.67 1.67 0 01.589-.41c.257-.1.53-.118.796-.042.401.114.745.368 1.016.737.248.337.434.769.561 1.287.23.934.27 2.163.115 3.645l.053.04.026.019c.757.576 1.284 1.397 1.563 2.35.435 1.487.216 3.155-.534 4.088l-.018.021.002.003c.417.762.67 1.567.724 2.4l.002.03c.064 1.065-.2 2.137-.814 3.19l-.007.01.01.024c.472 1.157.62 2.322.438 3.486l-.006.039a.651.651 0 01-.747.536.648.648 0 01-.54-.742c.167-1.033.01-2.069-.48-3.123a.643.643 0 01.04-.617l.004-.006c.604-.924.854-1.83.8-2.72-.046-.779-.325-1.544-.8-2.273a.644.644 0 01.18-.886l.009-.006c.243-.159.467-.565.58-1.12a4.229 4.229 0 00-.095-1.974c-.205-.7-.58-1.284-1.105-1.683-.595-.454-1.383-.673-2.38-.61a.653.653 0 01-.632-.371c-.314-.665-.772-1.141-1.343-1.436a3.288 3.288 0 00-1.772-.332c-1.245.099-2.343.801-2.67 1.686a.652.652 0 01-.61.425c-1.067.002-1.893.252-2.497.703-.522.39-.878.935-1.066 1.588a4.07 4.07 0 00-.068 1.886c.112.558.331 1.02.582 1.269l.008.007c.212.207.257.53.109.785-.36.622-.629 1.549-.673 2.44-.05 1.018.186 1.902.719 2.536l.016.019a.643.643 0 01.095.69c-.576 1.236-.753 2.252-.562 3.052a.652.652 0 01-1.269.298c-.243-1.018-.078-2.184.473-3.498l.014-.035-.008-.012a4.339 4.339 0 01-.598-1.309l-.005-.019a5.764 5.764 0 01-.177-1.785c.044-.91.278-1.842.622-2.59l.012-.026-.002-.002c-.293-.418-.51-.953-.63-1.545l-.005-.024a5.352 5.352 0 01.093-2.49c.262-.915.777-1.701 1.536-2.269.06-.045.123-.09.186-.132-.159-1.493-.119-2.73.112-3.67.127-.518.314-.95.562-1.287.27-.368.614-.622 1.015-.737.266-.076.54-.059.797.042zm4.116 9.09c.936 0 1.8.313 2.446.855.63.527 1.005 1.235 1.005 1.94 0 .888-.406 1.58-1.133 2.022-.62.375-1.451.557-2.403.557-1.009 0-1.871-.259-2.493-.734-.617-.47-.963-1.13-.963-1.845 0-.707.398-1.417 1.056-1.946.668-.537 1.55-.849 2.485-.849zm0 .896a3.07 3.07 0 00-1.916.65c-.461.37-.722.835-.722 1.25 0 .428.21.829.61 1.134.455.347 1.124.548 1.943.548.799 0 1.473-.147 1.932-.426.463-.28.7-.686.7-1.257 0-.423-.246-.89-.683-1.256-.484-.405-1.14-.643-1.864-.643zm.662 1.21l.004.004c.12.151.095.37-.056.49l-.292.23v.446a.375.375 0 01-.376.373.375.375 0 01-.376-.373v-.46l-.271-.218a.347.347 0 01-.052-.49.353.353 0 01.494-.051l.215.172.22-.174a.353.353 0 01.49.051zm-5.04-1.919c.478 0 .867.39.867.871a.87.87 0 01-.868.871.87.87 0 01-.867-.87.87.87 0 01.867-.872zm8.706 0c.48 0 .868.39.868.871a.87.87 0 01-.868.871.87.87 0 01-.867-.87.87.87 0 01.867-.872zM7.44 2.3l-.003.002a.659.659 0 00-.285.238l-.005.006c-.138.189-.258.467-.348.832-.17.692-.216 1.631-.124 2.782.43-.128.899-.208 1.404-.237l.01-.001.019-.034c.046-.082.095-.161.148-.239.123-.771.022-1.692-.253-2.444-.134-.364-.297-.65-.453-.813a.628.628 0 00-.107-.09L7.44 2.3zm9.174.04l-.002.001a.628.628 0 00-.107.09c-.156.163-.32.45-.453.814-.29.794-.387 1.776-.23 2.572l.058.097.008.014h.03a5.184 5.184 0 011.466.212c.086-1.124.038-2.043-.128-2.722-.09-.365-.21-.643-.349-.832l-.004-.006a.659.659 0 00-.285-.239h-.004z"></path></svg>

frontend/src/assets/providers/openai.svg

@@ -0,0 +1 @@
+<svg fill="currentColor" fill-rule="evenodd" height="1em" style="flex:none;line-height:1" viewBox="0 0 24 24" width="1em" xmlns="http://www.w3.org/2000/svg"><title>OpenAI</title><path d="M9.205 8.658v-2.26c0-.19.072-.333.238-.428l4.543-2.616c.619-.357 1.356-.523 2.117-.523 2.854 0 4.662 2.212 4.662 4.566 0 .167 0 .357-.024.547l-4.71-2.759a.797.797 0 00-.856 0l-5.97 3.473zm10.609 8.8V12.06c0-.333-.143-.57-.429-.737l-5.97-3.473 1.95-1.118a.433.433 0 01.476 0l4.543 2.617c1.309.76 2.189 2.378 2.189 3.948 0 1.808-1.07 3.473-2.76 4.163zM7.802 12.703l-1.95-1.142c-.167-.095-.239-.238-.239-.428V5.899c0-2.545 1.95-4.472 4.591-4.472 1 0 1.927.333 2.712.928L8.23 5.067c-.285.166-.428.404-.428.737v6.898zM12 15.128l-2.795-1.57v-3.33L12 8.658l2.795 1.57v3.33L12 15.128zm1.796 7.23c-1 0-1.927-.332-2.712-.927l4.686-2.712c.285-.166.428-.404.428-.737v-6.898l1.974 1.142c.167.095.238.238.238.428v5.233c0 2.545-1.974 4.472-4.614 4.472zm-5.637-5.303l-4.544-2.617c-1.308-.761-2.188-2.378-2.188-3.948A4.482 4.482 0 014.21 6.327v5.423c0 .333.143.571.428.738l5.947 3.449-1.95 1.118a.432.432 0 01-.476 0zm-.262 3.9c-2.688 0-4.662-2.021-4.662-4.519 0-.19.024-.38.047-.57l4.686 2.71c.286.167.571.167.856 0l5.97-3.448v2.26c0 .19-.07.333-.237.428l-4.543 2.616c-.619.357-1.356.523-2.117.523zm5.899 2.83a5.947 5.947 0 005.827-4.756C22.287 18.339 24 15.84 24 13.296c0-1.665-.713-3.282-1.998-4.448.119-.5.19-.999.19-1.498 0-3.401-2.759-5.947-5.946-5.947-.642 0-1.26.095-1.88.31A5.962 5.962 0 0010.205 0a5.947 5.947 0 00-5.827 4.757C1.713 5.447 0 7.945 0 10.49c0 1.666.713 3.283 1.998 4.448-.119.5-.19 1-.19 1.499 0 3.401 2.759 5.946 5.946 5.946.642 0 1.26-.095 1.88-.309a5.96 5.96 0 004.162 1.713z"></path></svg>

frontend/src/assets/providers/qwen.svg

@@ -0,0 +1 @@
+<svg height="1em" style="flex:none;line-height:1" viewBox="0 0 24 24" width="1em" xmlns="http://www.w3.org/2000/svg"><title>Qwen</title><path d="M12.604 1.34c.393.69.784 1.382 1.174 2.075a.18.18 0 00.157.091h5.552c.174 0 .322.11.446.327l1.454 2.57c.19.337.24.478.024.837-.26.43-.513.864-.76 1.3l-.367.658c-.106.196-.223.28-.04.512l2.652 4.637c.172.301.111.494-.043.77-.437.785-.882 1.564-1.335 2.34-.159.272-.352.375-.68.37-.777-.016-1.552-.01-2.327.016a.099.099 0 00-.081.05 575.097 575.097 0 01-2.705 4.74c-.169.293-.38.363-.725.364-.997.003-2.002.004-3.017.002a.537.537 0 01-.465-.271l-1.335-2.323a.09.09 0 00-.083-.049H4.982c-.285.03-.553-.001-.805-.092l-1.603-2.77a.543.543 0 01-.002-.54l1.207-2.12a.198.198 0 000-.197 550.951 550.951 0 01-1.875-3.272l-.79-1.395c-.16-.31-.173-.496.095-.965.465-.813.927-1.625 1.387-2.436.132-.234.304-.334.584-.335a338.3 338.3 0 012.589-.001.124.124 0 00.107-.063l2.806-4.895a.488.488 0 01.422-.246c.524-.001 1.053 0 1.583-.006L11.704 1c.341-.003.724.032.9.34zm-3.432.403a.06.06 0 00-.052.03L6.254 6.788a.157.157 0 01-.135.078H3.253c-.056 0-.07.025-.041.074l5.81 10.156c.025.042.013.062-.034.063l-2.795.015a.218.218 0 00-.2.116l-1.32 2.31c-.044.078-.021.118.068.118l5.716.008c.046 0 .08.02.104.061l1.403 2.454c.046.081.092.082.139 0l5.006-8.76.783-1.382a.055.055 0 01.096 0l1.424 2.53a.122.122 0 00.107.062l2.763-.02a.04.04 0 00.035-.02.041.041 0 000-.04l-2.9-5.086a.108.108 0 010-.113l.293-.507 1.12-1.977c.024-.041.012-.062-.035-.062H9.2c-.059 0-.073-.026-.043-.077l1.434-2.505a.107.107 0 000-.114L9.225 1.774a.06.06 0 00-.053-.031zm6.29 8.02c.046 0 .058.02.034.06l-.832 1.465-2.613 4.585a.056.056 0 01-.05.029.058.058 0 01-.05-.029L8.498 9.841c-.02-.034-.01-.052.028-.054l.216-.012 6.722-.012z" fill="url(#lobe-icons-qwen-_R_0_)" fill-rule="nonzero"></path><defs><linearGradient id="lobe-icons-qwen-_R_0_" x1="0%" x2="100%" y1="0%" y2="0%"><stop offset="0%" stop-color="#6336E7" stop-opacity=".84"></stop><stop offset="100%" stop-color="#6F69F7" stop-opacity=".84"></stop></linearGradient></defs></svg>

frontend/src/assets/providers/siliconflow.svg

@@ -0,0 +1 @@
+<svg height="1em" style="flex:none;line-height:1" viewBox="0 0 24 24" width="1em" xmlns="http://www.w3.org/2000/svg"><title>SiliconCloud</title><path clip-rule="evenodd" d="M22.956 6.521H12.522c-.577 0-1.044.468-1.044 1.044v3.13c0 .577-.466 1.044-1.043 1.044H1.044c-.577 0-1.044.467-1.044 1.044v4.174C0 17.533.467 18 1.044 18h10.434c.577 0 1.044-.467 1.044-1.043v-3.13c0-.578.466-1.044 1.043-1.044h9.391c.577 0 1.044-.467 1.044-1.044V7.565c0-.576-.467-1.044-1.044-1.044z" fill="#6E29F6" fill-rule="evenodd"></path></svg>

frontend/src/assets/providers/stepfun.svg

@@ -0,0 +1 @@
+<svg height="1em" style="flex:none;line-height:1" viewBox="0 0 24 24" width="1em" xmlns="http://www.w3.org/2000/svg"><title>Stepfun</title><path d="M22.012 0h1.032v.927H24v.968h-.956V3.78h-1.032V1.896h-1.878v-.97h1.878V0zM2.6 12.371V1.87h.969v10.502h-.97zm10.423.66h10.95v.918h-6.208v9.579h-4.742V13.03zM5.629 3.333v12.356H0v4.51h10.386V8L20.859 8l-.003-4.668-15.227.001z" fill="url(#lobe-icons-stepfun-_R_0_)" fill-rule="evenodd"></path><defs><linearGradient gradientUnits="userSpaceOnUse" id="lobe-icons-stepfun-_R_0_" x1="1.646" x2="18.342" y1="1.916" y2="22.091"><stop stop-color="#01A9FF"></stop><stop offset="1" stop-color="#0160FF"></stop></linearGradient></defs></svg>

frontend/src/assets/providers/volcengine.svg

@@ -0,0 +1 @@
+<svg height="1em" style="flex:none;line-height:1" viewBox="0 0 24 24" width="1em" xmlns="http://www.w3.org/2000/svg"><title>Volcengine</title><path d="M19.44 10.153l-2.936 11.586a.215.215 0 00.214.261h5.87a.215.215 0 00.214-.261l-2.95-11.586a.214.214 0 00-.412 0zM3.28 12.778l-2.275 8.96A.214.214 0 001.22 22h4.532a.212.212 0 00.214-.165.214.214 0 000-.097l-2.276-8.96a.214.214 0 00-.41 0z" fill="#00E5E5"></path><path d="M7.29 5.359L3.148 21.738a.215.215 0 00.203.261h8.29a.214.214 0 00.215-.261L7.7 5.358a.214.214 0 00-.41 0z" fill="#006EFF"></path><path d="M14.44.15a.214.214 0 00-.41 0L8.366 21.739a.214.214 0 00.214.261H19.9a.216.216 0 00.171-.078.214.214 0 00.044-.183L14.439.15z" fill="#006EFF"></path><path d="M10.278 7.741L6.685 21.736a.214.214 0 00.214.264h7.17a.215.215 0 00.214-.264L10.688 7.741a.214.214 0 00-.41 0z" fill="#00E5E5"></path></svg>

frontend/src/assets/providers/zhipu.svg

@@ -0,0 +1 @@
+<svg height="1em" style="flex:none;line-height:1" viewBox="0 0 24 24" width="1em" xmlns="http://www.w3.org/2000/svg"><title>Zhipu</title><path d="M11.991 23.503a.24.24 0 00-.244.248.24.24 0 00.244.249.24.24 0 00.245-.249.24.24 0 00-.22-.247l-.025-.001zM9.671 5.365a1.697 1.697 0 011.099 2.132l-.071.172-.016.04-.018.054c-.07.16-.104.32-.104.498-.035.71.47 1.279 1.186 1.314h.366c1.309.053 2.338 1.173 2.286 2.523-.052 1.332-1.152 2.38-2.478 2.327h-.174c-.715.018-1.274.64-1.239 1.368 0 .124.018.23.053.337.209.373.54.658.96.8.75.23 1.517-.125 1.9-.782l.018-.035c.402-.64 1.17-.96 1.92-.711.854.284 1.378 1.226 1.099 2.167a1.661 1.661 0 01-2.077 1.102 1.711 1.711 0 01-.907-.711l-.017-.035c-.2-.323-.463-.58-.851-.711l-.056-.018a1.646 1.646 0 00-1.954.746 1.66 1.66 0 01-1.065.764 1.677 1.677 0 01-1.989-1.279c-.209-.906.332-1.83 1.257-2.043a1.51 1.51 0 01.296-.035h.018c.68-.071 1.151-.622 1.116-1.333a1.307 1.307 0 00-.227-.693 2.515 2.515 0 01-.366-1.403 2.39 2.39 0 01.366-1.208c.14-.195.21-.444.227-.693.018-.71-.506-1.261-1.186-1.332l-.07-.018a1.43 1.43 0 01-.299-.07l-.05-.019a1.7 1.7 0 01-1.047-2.114 1.68 1.68 0 012.094-1.101zm-5.575 10.11c.26-.264.639-.367.994-.27.355.096.633.379.728.74.095.362-.007.748-.267 1.013-.402.41-1.053.41-1.455 0a1.062 1.062 0 010-1.482zm14.845-.294c.359-.09.738.024.992.297.254.274.344.665.237 1.025-.107.36-.396.634-.756.718-.551.128-1.1-.22-1.23-.781a1.05 1.05 0 01.757-1.26zm-.064-4.39c.314.32.49.753.49 1.206 0 .452-.176.886-.49 1.206-.315.32-.74.5-1.185.5-.444 0-.87-.18-1.184-.5a1.727 1.727 0 010-2.412 1.654 1.654 0 012.369 0zm-11.243.163c.364.484.447 1.128.218 1.691a1.665 1.665 0 01-2.188.923c-.855-.36-1.26-1.358-.907-2.228a1.68 1.68 0 011.33-1.038c.593-.08 1.183.169 1.547.652zm11.545-4.221c.368 0 .708.2.892.524.184.324.184.724 0 1.048a1.026 1.026 0 01-.892.524c-.568 0-1.03-.47-1.03-1.048 0-.579.462-1.048 1.03-1.048zm-14.358 0c.368 0 .707.2.891.524.184.324.184.724 0 1.048a1.026 1.026 0 01-.891.524c-.569 0-1.03-.47-1.03-1.048 0-.579.461-1.048 1.03-1.048zm10.031-1.475c.925 0 1.675.764 1.675 1.706s-.75 1.705-1.675 1.705-1.674-.763-1.674-1.705c0-.942.75-1.706 1.674-1.706zm-2.626-.684c.362-.082.653-.356.761-.718a1.062 1.062 0 00-.238-1.028 1.017 1.017 0 00-.996-.294c-.547.14-.881.7-.752 1.257.13.558.675.907 1.225.783zm0 16.876c.359-.087.644-.36.75-.72a1.062 1.062 0 00-.237-1.019 1.018 1.018 0 00-.985-.301 1.037 1.037 0 00-.762.717c-.108.361-.017.754.239 1.028.245.263.606.377.953.305l.043-.01zM17.19 3.5a.631.631 0 00.628-.64c0-.355-.279-.64-.628-.64a.631.631 0 00-.628.64c0 .355.28.64.628.64zm-10.38 0a.631.631 0 00.628-.64c0-.355-.28-.64-.628-.64a.631.631 0 00-.628.64c0 .355.279.64.628.64zm-5.182 7.852a.631.631 0 00-.628.64c0 .354.28.639.628.639a.63.63 0 00.627-.606l.001-.034a.62.62 0 00-.628-.64zm5.182 9.13a.631.631 0 00-.628.64c0 .355.279.64.628.64a.631.631 0 00.628-.64c0-.355-.28-.64-.628-.64zm10.38.018a.631.631 0 00-.628.64c0 .355.28.64.628.64a.631.631 0 00.628-.64c0-.355-.279-.64-.628-.64zm5.182-9.148a.631.631 0 00-.628.64c0 .354.279.639.628.639a.631.631 0 00.628-.64c0-.355-.28-.64-.628-.64zm-.384-4.992a.24.24 0 00.244-.249.24.24 0 00-.244-.249.24.24 0 00-.244.249c0 .142.122.249.244.249zM11.991.497a.24.24 0 00.245-.248A.24.24 0 0011.99 0a.24.24 0 00-.244.249c0 .133.108.236.223.247l.021.001zM2.011 6.36a.24.24 0 00.245-.249.24.24 0 00-.244-.249.24.24 0 00-.244.249.24.24 0 00.244.249zm0 11.263a.24.24 0 00-.243.248.24.24 0 00.244.249.24.24 0 00.244-.249.252.252 0 00-.244-.248zm19.995-.018a.24.24 0 00-.245.248.24.24 0 00.245.25.24.24 0 00.244-.25.252.252 0 00-.244-.248z" fill="#3859FF" fill-rule="nonzero"></path></svg>

frontend/src/components/common/StatusBar.vue

@@ -39,11 +39,12 @@ const saveStatusColor = computed(() => {
 
 const indexStatusText = computed(() => {
   const s = settingsStore.indexStatus.status
-  return s === 'idle' ? '索引就绪' : s === 'indexing' ? `索引中 (${settingsStore.indexStatus.pending_jobs})` : '索引错误'
+  return s === 'unknown' ? '索引状态未获取' : s === 'idle' ? '索引就绪' : s === 'indexing' ? `索引中 (${settingsStore.indexStatus.pending_jobs})` : '索引错误'
 })
 
 const aiCoreStatusText = computed(() => {
   const map: Record<string, string> = {
+    unknown: 'AI Core 状态未获取',
     starting: 'AI Core 启动中',
     running: 'AI Core 运行中',
     stopped: 'AI Core 已停止',

frontend/src/contracts/index.ts

@@ -207,8 +207,8 @@ export interface PermissionRequest {
 }
 
 export interface TokenUsage {
-  input_tokens: number
+  input_tokens?: number
-  output_tokens: number
+  output_tokens?: number
   total_tokens: number
 }
 
@@ -416,6 +416,39 @@ export interface ProviderPreset {
   base_url: string
   default_credential_id?: string | null
   requires_credential: boolean
+  logo_id?: string
+  description?: string
+  capabilities?: string[]
+}
+
+export type ProviderUpdateRequest = Partial<Omit<ProviderConfig, 'provider_id' | 'credential_id' | 'base_url'>> & {
+  credential_id?: string | null
+  base_url?: string | null
+}
+
+export type RoutingCapability = 'embedding' | 'transcription' | 'speaker_matching'
+
+export interface ModelBinding {
+  provider_id: string
+  model: string
+  endpoint: string
+  dimensions?: number | null
+}
+
+export interface ModelRoutingConfig {
+  version: number
+  embedding: ModelBinding | null
+  transcription: ModelBinding | null
+  speaker_matching: ModelBinding | null
+}
+
+export interface ModelRoutingResponse {
+  config: ModelRoutingConfig
+  local_backends: Array<{
+    capability: RoutingCapability
+    status: 'placeholder' | 'not_installed' | 'ready'
+    message: string
+  }>
 }
 
 // ============ Tasks ============
@@ -429,11 +462,11 @@ export interface TaskItem {
   title: string
   description?: string
   status: TaskStatus
-  priority: TaskPriority
+  priority?: TaskPriority
   due_date?: string
   note_id?: string
   note_title?: string
-  source: TaskSource
+  source?: TaskSource
   created_at: string
   updated_at: string
 }
@@ -454,12 +487,12 @@ export interface ThemeConfig {
 // ============ Index ============
 
 export interface IndexStatus {
-  status: 'idle' | 'indexing' | 'error'
+  status: 'unknown' | 'idle' | 'indexing' | 'error'
   pending_jobs: number
-  total_notes: number
+  total_notes: number | null
-  total_blocks: number
+  total_blocks: number | null
-  fts_enabled: boolean
+  fts_enabled?: boolean
-  vector_enabled: boolean
+  vector_enabled?: boolean
   embedding_model?: string
   reranker_model?: string
   last_indexed_at?: string
@@ -494,7 +527,7 @@ export type SaveStatus =
   | 'external_changed'
   | 'conflict'
 
-export type AiCoreStatus = 'starting' | 'running' | 'stopped' | 'error'
+export type AiCoreStatus = 'unknown' | 'starting' | 'running' | 'stopped' | 'error'
 
 // ============ FastAPI wire contracts ============
 // UI view models above may contain presentation-only fields. Services must use
@@ -721,6 +754,9 @@ export interface ApiProviderPreset {
   base_url: string
   default_credential_id?: string | null
   requires_credential: boolean
+  logo_id?: string
+  description?: string
+  capabilities?: string[]
 }
 
 export interface ApiModelInfo {
@@ -741,6 +777,8 @@ export interface ApiTask {
 }
 
 export interface ApiIndexStatus {
+  total_notes: number
+  total_blocks: number
   status: 'idle' | 'queued' | 'running' | 'failed'
   pending_jobs: number
   active_job_id?: string | null

frontend/src/features/agent/AgentView.vue

@@ -5,7 +5,8 @@ import { useAgentStore } from '@/stores/agent'
 import { useProviderStore } from '@/stores/provider'
 import { useSkillStore } from '@/stores/skill'
 import type { AgentEvent } from '@/contracts'
-import { eventLabel, localizeDetails, permissionLabel, runStatusLabel, toolDescription, toolLabel } from './labels'
+import { eventLabel, localizeDetails, permissionLabel, runStatusLabel, toolLabel } from './labels'
+import ToolOption from './ToolOption.vue'
 
 const route = useRoute()
 const router = useRouter()
@@ -14,7 +15,7 @@ const providerStore = useProviderStore()
 const skillStore = useSkillStore()
 const pageError = ref('')
 const form = reactive({
-  input: '', provider_id: 'mock', model: 'mock-1', skill_id: '', max_steps: 10,
+  input: '', provider_id: '', model: '', skill_id: '', max_steps: 10,
   tool_timeout_seconds: 30, run_timeout_seconds: 300, token_budget: 8000,
   allow_network: false, max_concurrent_tools: 1, allowed_tools: [] as string[],
 })
@@ -25,7 +26,7 @@ const isNewRun = computed(() => !route.params.runId)
 onMounted(async () => {
   try {
     await Promise.all([providerStore.loadProviders(), skillStore.loadSkills(), agentStore.loadTools()])
-    await providerStore.loadModels(form.provider_id)
+    form.provider_id = providerStore.defaultProviderId
   } catch (error) { pageError.value = error instanceof Error ? error.message : '智能体配置加载失败' }
 })
 
@@ -35,7 +36,10 @@ watch(() => route.params.runId, async (runId) => {
 }, { immediate: true })
 
 watch(() => form.provider_id, async (providerId) => {
-  try { await providerStore.loadModels(providerId); form.model = models.value[0]?.model_id ?? '' } catch { /* page keeps current selection */ }
+  form.model = providerStore.providers.find(p => p.provider_id === providerId)?.default_model ?? ''
+  if (!providerId) return
+  try { await providerStore.loadModels(providerId) }
+  catch (error) { if (form.provider_id === providerId) pageError.value = error instanceof Error ? error.message : '模型列表加载失败，请手动填写模型 ID。' }
 })
 
 function toggleTool(name: string) {
@@ -47,6 +51,7 @@ function toggleTool(name: string) {
 async function createRun() {
   pageError.value = ''
   try {
+    if (!form.provider_id || !form.model.trim()) throw new Error('请选择提供商并填写模型 ID。')
     const run = await agentStore.createRun({
       input: form.input, provider_id: form.provider_id, model: form.model,
       skill_id: form.skill_id || undefined, allowed_tools: form.allowed_tools,
@@ -71,12 +76,12 @@ function eventText(event: AgentEvent) {
   <section class="feature-page agent-page">
     <header class="feature-header"><div><h1>{{ isNewRun ? '创建智能体运行' : '智能体执行轨迹' }}</h1><p>配置执行边界，并实时查看模型、工具和权限事件。</p></div>
       <button v-if="!isNewRun" class="button-secondary" @click="router.push({ name: 'agent' })">新建运行</button></header>
-    <div v-if="pageError || agentStore.error" class="error-banner">{{ pageError || agentStore.error }}</div>
+    <div v-if="pageError || agentStore.error || providerStore.error" class="error-banner">{{ pageError || agentStore.error || providerStore.error }}</div>
     <form v-if="isNewRun" class="panel run-form" @submit.prevent="createRun">
       <div class="field"><label>任务</label><textarea v-model="form.input" class="textarea" required placeholder="描述希望智能体完成的任务" /></div>
       <div class="form-grid">
         <div class="field"><label>模型提供商</label><select v-model="form.provider_id" class="select"><option v-for="p in providerStore.enabledProviders" :key="p.provider_id" :value="p.provider_id">{{ p.name }}</option></select></div>
-        <div class="field"><label>模型</label><select v-model="form.model" class="select"><option v-for="m in models" :key="m.model_id" :value="m.model_id">{{ m.name }}</option></select></div>
+        <div class="field"><label>模型</label><input v-model="form.model" class="input" list="agent-models" placeholder="填写模型 ID" required /><datalist id="agent-models"><option v-for="m in models" :key="m.model_id" :value="m.model_id">{{ m.name }}</option></datalist></div>
         <div class="field"><label>技能</label><select v-model="form.skill_id" class="select"><option value="">不使用技能</option><option v-for="s in skillStore.readySkills" :key="s.skill_id" :value="s.skill_id">{{ s.name }}</option></select></div>
         <div class="field"><label>最大步骤</label><input v-model.number="form.max_steps" class="input" type="number" min="1" max="100" /></div>
         <div class="field"><label>工具超时（秒）</label><input v-model.number="form.tool_timeout_seconds" class="input" type="number" min="1" /></div>
@@ -84,9 +89,9 @@ function eventText(event: AgentEvent) {
         <div class="field"><label>令牌预算</label><input v-model.number="form.token_budget" class="input" type="number" min="1" /></div>
         <div class="field"><label>最大并发工具</label><input v-model.number="form.max_concurrent_tools" class="input" type="number" min="1" /></div>
       </div>
-      <div class="field"><label>允许使用的工具</label><div class="tool-grid"><label v-for="tool in agentStore.tools" :key="tool.name" class="tool-option"><input type="checkbox" :checked="form.allowed_tools.includes(tool.name)" @change="toggleTool(tool.name)" /><span><strong>{{ toolLabel(tool.name) }}</strong><code>{{ tool.name }}</code><small>{{ toolDescription(tool.name, tool.description) }}</small></span></label></div></div>
+      <div class="field"><label>允许使用的工具</label><div class="tool-grid"><ToolOption v-for="tool in agentStore.tools" :key="tool.name" :name="tool.name" :description="tool.description" :selected="form.allowed_tools.includes(tool.name)" @toggle="toggleTool" /></div></div>
       <label class="network"><input v-model="form.allow_network" type="checkbox" /> 允许本次运行调用网络工具</label>
-      <div class="inline-actions"><button class="button-primary" :disabled="agentStore.isCreating || !form.input.trim()">{{ agentStore.isCreating ? '创建中…' : '创建并运行' }}</button></div>
+      <div class="inline-actions"><button class="button-primary" :disabled="agentStore.isCreating || !form.input.trim() || !form.provider_id || !form.model.trim()">{{ agentStore.isCreating ? '创建中…' : '创建并运行' }}</button></div>
     </form>
 
     <div v-else class="trace-layout">
@@ -110,12 +115,7 @@ function eventText(event: AgentEvent) {
 <style scoped>
 .agent-page > * { width: min(100%, 1080px); margin-inline: auto; }
 .run-form { display: grid; gap: var(--space-xl); }
-.tool-grid { display: grid; grid-template-columns: repeat(auto-fit, minmax(230px, 1fr)); gap: var(--space-sm); }
+.tool-grid { display: grid; align-items: start; grid-template-columns: repeat(auto-fit, minmax(230px, 1fr)); gap: var(--space-sm); }
-.tool-option { display: flex; gap: var(--space-sm); padding: var(--space-md); border: 1px solid var(--color-border-default); border-radius: var(--radius-md); background: var(--color-surface-primary); cursor: pointer; transition: border-color var(--motion-fast), background-color var(--motion-fast), transform var(--motion-fast), box-shadow var(--motion-fast); }
-.tool-option:hover { border-color: var(--color-accent-secondary); transform: translateY(-1px); box-shadow: var(--shadow-sm); }
-.tool-option:has(input:checked) { border-color: var(--color-accent-primary); background: var(--color-accent-soft); box-shadow: 0 0 0 2px color-mix(in srgb, var(--color-accent-primary) 10%, transparent); }
-.tool-option small { display: block; color: var(--color-text-secondary); }
-.tool-option code { display: block; margin: 2px 0; color: var(--color-text-tertiary); font-size: var(--font-size-xs); }
 .network { display: flex; gap: var(--space-sm); }
 .trace-layout { display: grid; gap: var(--space-lg); }
 .run-summary, .event-head { display: flex; align-items: center; justify-content: space-between; gap: var(--space-md); }

frontend/src/features/agent/ToolOption.spec.ts

@@ -0,0 +1,19 @@
+// @vitest-environment happy-dom
+import { mount } from '@vue/test-utils'
+import { expect, it } from 'vitest'
+import ToolOption from './ToolOption.vue'
+
+it('shows Chinese summaries, preserves raw metadata and emits the original tool ID', async () => {
+  const name = 'mcp.9ca7ee21603a.web_search'
+  const description = 'Search the web. query: string. ' + 'Full provider instructions. '.repeat(40)
+  const wrapper = mount(ToolOption, { props: { name, description, selected: false } })
+  expect(wrapper.get('strong').text()).toBe('网页搜索')
+  expect(wrapper.get('code').text()).toBe(name)
+  expect(wrapper.get('.tool-summary').text()).toContain('搜索关键词')
+  expect(wrapper.get('details').attributes('open')).toBeUndefined()
+  expect(wrapper.get('details p').element.textContent).toBe(description)
+  await wrapper.get('summary').trigger('click')
+  expect(wrapper.emitted('toggle')).toBeUndefined()
+  await wrapper.get('input').setValue(true)
+  expect(wrapper.emitted('toggle')).toEqual([[name]])
+})

frontend/src/features/agent/ToolOption.vue

@@ -0,0 +1,40 @@
+<script setup lang="ts">
+import { computed } from 'vue'
+import { toolDescription, toolLabel } from './labels'
+
+const props = defineProps<{ name: string; description: string; selected: boolean }>()
+const emit = defineEmits<{ toggle: [name: string] }>()
+const summary = computed(() => toolDescription(props.name, props.description))
+const showOriginal = computed(() => props.description.length > 0)
+</script>
+
+<template>
+  <article class="tool-choice" :class="{ selected }">
+    <label class="tool-selection">
+      <input type="checkbox" :checked="selected" @change="emit('toggle', name)" />
+      <span class="tool-copy">
+        <strong>{{ toolLabel(name) }}</strong>
+        <code>{{ name }}</code>
+        <small class="tool-summary">{{ summary }}</small>
+      </span>
+    </label>
+    <details v-if="showOriginal" class="tool-original">
+      <summary>查看服务原文与参数</summary>
+      <p>{{ description }}</p>
+    </details>
+  </article>
+</template>
+
+<style scoped>
+.tool-choice { min-width: 0; padding: var(--space-md); border: 1px solid var(--color-border-default); border-radius: var(--radius-md); background: var(--color-surface-primary); }
+.tool-choice.selected { border-color: var(--color-accent-primary); background: var(--color-accent-soft); }
+.tool-selection { display: flex; align-items: flex-start; gap: var(--space-sm); cursor: pointer; }
+.tool-selection input { flex-shrink: 0; margin-top: 4px; }
+.tool-copy { min-width: 0; overflow-wrap: anywhere; }
+.tool-copy strong, .tool-copy code, .tool-summary { display: block; }
+.tool-copy code { margin: 3px 0; color: var(--color-text-tertiary); font-size: var(--font-size-xs); }
+.tool-summary { color: var(--color-text-secondary); line-height: 1.6; display: -webkit-box; -webkit-box-orient: vertical; -webkit-line-clamp: 3; overflow: hidden; }
+.tool-original { margin-top: var(--space-sm); font-size: var(--font-size-xs); }
+.tool-original summary { cursor: pointer; color: var(--color-text-secondary); }
+.tool-original p { white-space: pre-wrap; overflow-wrap: anywhere; max-height: 240px; overflow: auto; margin-top: var(--space-sm); user-select: text; }
+</style>

frontend/src/features/agent/labels.spec.ts

@@ -9,6 +9,21 @@ import {
 } from './labels'
 
 describe('智能体页面中文标签', () => {
+  it('按 MCP 远程工具名匹配中文，不依赖服务器 ID', () => {
+    for (const server of ['9ca7ee21603a', 'another-server']) {
+      expect(toolLabel(`mcp.${server}.web_search`)).toBe('网页搜索')
+      expect(toolLabel(`mcp.${server}.understand_image`)).toBe('图像理解')
+      expect(toolDescription(`mcp.${server}.web_search`, 'Search the web')).toContain('搜索关键词')
+    }
+    expect(toolLabel('text.uppercase')).toBe('文本转大写')
+    expect(toolDescription('text.uppercase', 'Convert input text to uppercase.')).toContain('大写')
+  })
+
+  it('保留服务端中文，未知工具不编造翻译或套用内置工具语义', () => {
+    expect(toolDescription('mcp.server.web_search', '仅搜索指定站点。')).toBe('仅搜索指定站点。')
+    expect(toolDescription('mcp.server.custom_action', 'Private action')).toContain('暂无中文说明')
+    expect(toolLabel('mcp.server.notes.delete')).toBe('MCP 工具 · notes.delete')
+  })
   it('转换运行状态和事件名称', () => {
     expect(runStatusLabel('waiting_permission')).toBe('等待授权')
     expect(eventLabel('ToolCall')).toBe('调用工具')

frontend/src/features/agent/labels.ts

@@ -42,6 +42,7 @@ const toolLabels: Record<string, string> = {
   'tasks.list': '列出任务',
   'attachments.read': '读取附件',
   'audio.transcribe': '音频转写',
+  'text.uppercase': '文本转大写',
 }
 
 const toolDescriptions: Record<string, string> = {
@@ -58,7 +59,25 @@ const toolDescriptions: Record<string, string> = {
   'tasks.update': '更新已有任务。',
   'tasks.list': '列出已持久化的任务。',
   'attachments.read': '读取由宿主管理的 UTF-8 附件。',
-  'audio.transcribe': '读取音频附件已有的宿主转写结果。',
+  'audio.transcribe': '将音频转写为文本，按模型路由使用 API 或本地后端。',
+  'text.uppercase': '将输入文本中的字母转换为大写。',
+}
+
+// MCP IDs contain a server-specific namespace. Localize the remote tool name
+// for presentation only; requests must keep using the complete original ID.
+const mcpTools: Record<string, { label: string; description: string }> = {
+  web_search: {
+    label: '网页搜索',
+    description: '搜索实时或外部网页信息。输入搜索关键词；结果包含标题、链接、摘要等信息。时效性问题可在关键词中加入日期，完整参数以服务原文为准。',
+  },
+  understand_image: {
+    label: '图像理解',
+    description: '根据提示词分析图片、描述内容或提取信息。输入分析要求和图片地址或本地路径；支持的格式与路径规则请查看服务原文。',
+  },
+}
+
+function mcpName(name: string): string | undefined {
+  return /^mcp\.[^.]+\.(.+)$/.exec(name)?.[1]
 }
 
 const permissionLabels: Record<string, string> = {
@@ -105,10 +124,17 @@ export function eventLabel(event: AgentEventType): string {
 }
 
 export function toolLabel(name: string): string {
+  const remote = mcpName(name)
+  if (remote) return mcpTools[remote]?.label ?? `MCP 工具 · ${remote}`
   return toolLabels[name] ?? name
 }
 
 export function toolDescription(name: string, fallback: string): string {
+  const remote = mcpName(name)
+  if (remote) {
+    if (/\p{Script=Han}/u.test(fallback)) return fallback
+    return mcpTools[remote]?.description ?? '暂无中文说明，请展开查看服务原文。'
+  }
   return toolDescriptions[name] ?? fallback
 }
 

frontend/src/features/chat/ChatView.spec.ts

@@ -0,0 +1,89 @@
+// @vitest-environment happy-dom
+import { beforeEach, expect, it, vi } from 'vitest'
+import { flushPromises, mount } from '@vue/test-utils'
+import { createPinia, setActivePinia } from 'pinia'
+import { useChatStore } from '@/stores/chat'
+import { useProviderStore } from '@/stores/provider'
+import { useSkillStore } from '@/stores/skill'
+import ChatView from './ChatView.vue'
+
+vi.mock('vue-router', () => ({ useRouter: () => ({ push: vi.fn() }) }))
+vi.mock('@/stores/editor', () => ({ useEditorStore: () => ({}) }))
+vi.mock('@/stores/workspace', () => ({ useWorkspaceStore: () => ({}) }))
+vi.mock('@/components/common/MarkdownContent.vue', () => ({ default: { template: '<div />' } }))
+
+beforeEach(() => {
+  setActivePinia(createPinia())
+  const providers = useProviderStore()
+  providers.providers = ['a', 'b'].map(id => ({
+    provider_id: id, provider_type: 'openai_compatible', name: id,
+    default_model: `${id}-default`, enabled: true, capabilities: { chat: true }, has_credential: false,
+  }))
+  providers.defaultProviderId = 'a'
+  vi.spyOn(providers, 'loadProviders').mockResolvedValue(undefined)
+  vi.spyOn(providers, 'loadModels').mockResolvedValue([])
+  vi.spyOn(useSkillStore(), 'loadSkills').mockResolvedValue(undefined)
+})
+
+it('preserves the selected provider and manual model after leaving and returning to chat', async () => {
+  const chat = useChatStore()
+  const first = mount(ChatView)
+  await flushPromises()
+  await first.get('select').setValue('b')
+  await first.get('input[list="chat-models"]').setValue('b-manual')
+  first.unmount()
+  const returned = mount(ChatView)
+  await flushPromises()
+  expect(chat.selectedProviderId).toBe('b')
+  expect(chat.selectedModel).toBe('b-manual')
+  expect(useProviderStore().loadModels).toHaveBeenLastCalledWith('b')
+  returned.unmount()
+})
+
+it.each(['missing', 'disabled', 'unselected'])('uses the default when the selected provider is %s', async state => {
+  const chat = useChatStore()
+  chat.selectedProviderId = state === 'unselected' ? '' : state === 'missing' ? 'deleted' : 'b'
+  chat.selectedModel = 'old-model'
+  if (state === 'disabled') useProviderStore().providers[1]!.enabled = false
+  const wrapper = mount(ChatView)
+  await flushPromises()
+  expect(chat.selectedProviderId).toBe('a')
+  expect(chat.selectedModel).toBe('a-default')
+  wrapper.unmount()
+})
+
+it('preserves the selection when provider discovery fails', async () => {
+  const chat = useChatStore()
+  chat.selectedProviderId = 'b'
+  chat.selectedModel = 'b-manual'
+  useProviderStore().error = 'offline'
+  const wrapper = mount(ChatView)
+  await flushPromises()
+  expect(chat.selectedProviderId).toBe('b')
+  expect(chat.selectedModel).toBe('b-manual')
+  expect(wrapper.get('.error-banner').text()).toBe('offline')
+  wrapper.unmount()
+})
+
+it.each(['providers', 'skills'])('ignores initialization after unmount while %s are loading', async source => {
+  const chat = useChatStore()
+  let finish!: () => void
+  const pending = new Promise<void>(resolve => { finish = resolve })
+  if (source === 'providers') vi.mocked(useProviderStore().loadProviders).mockReturnValueOnce(pending)
+  else vi.mocked(useSkillStore().loadSkills).mockReturnValueOnce(pending)
+  const first = mount(ChatView)
+  first.unmount()
+  finish()
+  await flushPromises()
+  expect(chat.selectedProviderId).toBe('')
+  expect(chat.selectedModel).toBe('')
+  expect(useProviderStore().loadModels).not.toHaveBeenCalled()
+
+  const returned = mount(ChatView)
+  await flushPromises()
+  expect(chat.selectedProviderId).toBe('a')
+  expect(chat.selectedModel).toBe('a-default')
+  await returned.get('textarea').setValue('hello')
+  expect(returned.get('button.button-primary').attributes('disabled')).toBeUndefined()
+  returned.unmount()
+})

frontend/src/features/chat/ChatView.vue

@@ -1,5 +1,5 @@
 <script setup lang="ts">
-import { computed, onMounted, ref, watch } from 'vue'
+import { computed, onBeforeUnmount, onMounted, ref, watch } from 'vue'
 import { useRouter } from 'vue-router'
 import type { Citation } from '@/contracts'
 import { useChatStore } from '@/stores/chat'
@@ -16,26 +16,37 @@ const workspaceStore = useWorkspaceStore()
 const editorStore = useEditorStore()
 const router = useRouter()
 const loadError = ref('')
+let disposed = false
+onBeforeUnmount(() => { disposed = true })
 
 const availableModels = computed(() => providerStore.modelsByProvider[chatStore.selectedProviderId] ?? [])
 
 onMounted(async () => {
   try {
     await Promise.all([providerStore.loadProviders(), skillStore.loadSkills()])
-    await providerStore.loadModels(chatStore.selectedProviderId)
+    if (disposed || providerStore.error) return
+    const selected = providerStore.enabledProviders.find(p => p.provider_id === chatStore.selectedProviderId)
+    if (!selected) {
+      chatStore.selectedProviderId = providerStore.defaultProviderId
+    } else {
+      await refreshModels(selected.provider_id)
+    }
   } catch (error) {
-    loadError.value = error instanceof Error ? error.message : '无法加载 AI 配置，当前展示本地数据。'
+    if (disposed) return
+    loadError.value = error instanceof Error ? error.message : '无法加载 AI 配置，请检查后端连接。'
   }
 })
 
+async function refreshModels(providerId: string) {
+  loadError.value = ''
+  if (!providerId) return
+  try { await providerStore.loadModels(providerId) }
+  catch (error) { if (!disposed && chatStore.selectedProviderId === providerId) loadError.value = error instanceof Error ? error.message : '模型列表加载失败，请手动填写模型 ID。' }
+}
+
 watch(() => chatStore.selectedProviderId, async (providerId) => {
-  try {
+  chatStore.selectedModel = providerStore.providers.find(p => p.provider_id === providerId)?.default_model ?? ''
-    await providerStore.loadModels(providerId)
+  await refreshModels(providerId)
-    const firstModel = providerStore.modelsByProvider[providerId]?.[0]
-    if (firstModel) chatStore.selectedModel = firstModel.model_id
-  } catch (error) {
-    loadError.value = error instanceof Error ? error.message : '模型列表加载失败'
-  }
 })
 
 function send() { void chatStore.sendMessage(chatStore.inputText) }
@@ -54,17 +65,12 @@ async function openCitation(citation: Citation) {
       <div class="field compact"><label>Provider</label><select v-model="chatStore.selectedProviderId" class="select">
         <option v-for="provider in providerStore.enabledProviders" :key="provider.provider_id" :value="provider.provider_id">{{ provider.name }}</option>
       </select></div>
-      <div class="field compact"><label>Model</label><select v-model="chatStore.selectedModel" class="select">
+      <div class="field compact"><label>模型 ID</label><input v-model="chatStore.selectedModel" class="input" list="chat-models" placeholder="填写模型 ID" /><datalist id="chat-models"><option v-for="model in availableModels" :key="model.model_id" :value="model.model_id">{{ model.name }}</option></datalist></div>
-        <option v-for="model in availableModels" :key="model.model_id" :value="model.model_id">{{ model.name }}</option>
+      <span class="subtle">知识库问答与技能请使用智能体；普通聊天尚未接入这些能力。</span>
-      </select></div>
-      <div class="field compact"><label>Skill</label><select v-model="chatStore.selectedSkillId" class="select">
-        <option :value="null">不使用 Skill</option><option v-for="skill in skillStore.enabledSkills" :key="skill.skill_id" :value="skill.skill_id">{{ skill.name }}</option>
-      </select></div>
-      <label class="rag-toggle"><input v-model="chatStore.useRag" type="checkbox" /> 使用知识库</label>
     </header>
-    <div v-if="loadError" class="error-banner chat-error">{{ loadError }}</div>
+    <div v-if="loadError || providerStore.error" class="error-banner chat-error">{{ loadError || providerStore.error }}</div>
     <main class="message-timeline">
-      <div v-if="!chatStore.messages.length" class="empty-state"><div><strong>开始一段知识对话</strong><p>可以直接提问，也可以打开 RAG 让模型基于当前 Vault 回答。</p></div></div>
+      <div v-if="!chatStore.messages.length" class="empty-state"><div><strong>开始一段知识对话</strong><p>请先配置模型提供商。聊天记录仅保留在本次页面会话中。</p></div></div>
       <article v-for="message in chatStore.messages" :key="message.message_id" class="message" :class="message.role">
         <div class="avatar">{{ message.role === 'user' ? '你' : 'AI' }}</div>
         <div class="message-body">
@@ -78,7 +84,7 @@ async function openCitation(citation: Citation) {
             </button>
           </div>
           <time>{{ new Date(message.created_at).toLocaleTimeString() }}</time>
-          <small v-if="message.usage" class="usage">Token {{ message.usage.total_tokens }}（输入 {{ message.usage.input_tokens }} / 输出 {{ message.usage.output_tokens }}）</small>
+          <small v-if="message.usage" class="usage">Token {{ message.usage.total_tokens }}<span v-if="message.usage.input_tokens !== undefined && message.usage.output_tokens !== undefined">（输入 {{ message.usage.input_tokens }} / 输出 {{ message.usage.output_tokens }}）</span></small>
         </div>
       </article>
     </main>
@@ -87,7 +93,7 @@ async function openCitation(citation: Citation) {
         @keydown.ctrl.enter.prevent="send" />
       <div class="composer-actions"><span class="subtle">回答可能包含错误，请核对 Citation。</span>
         <button v-if="chatStore.isStreaming" class="button-danger" @click="chatStore.stopGeneration">停止</button>
-        <button v-else class="button-primary" :disabled="!chatStore.inputText.trim()" @click="send">发送</button>
+        <button v-else class="button-primary" :disabled="!chatStore.inputText.trim() || !chatStore.selectedProviderId || !chatStore.selectedModel.trim()" @click="send">发送</button>
       </div>
     </footer>
   </section>

frontend/src/features/plugins/PluginsView.vue

@@ -27,6 +27,7 @@ async function uninstall(id: string, name: string) { if (!confirm(`卸载“${na
       <div v-if="pluginStore.selectedPlugin.dependent_skills?.length" class="notice-banner last-error">依赖此插件的 Skill：{{ pluginStore.selectedPlugin.dependent_skills.join('、') }}</div>
       <PluginMcpPanel :plugin="pluginStore.selectedPlugin" />
     </div>
+    <div v-else-if="!pluginStore.plugins.length" class="empty-state"><div><strong>{{ pluginStore.isLoading ? '正在加载…' : pluginStore.error ? '加载失败' : '尚未安装' }}</strong><button class="button-secondary" @click="pluginStore.loadPlugins">重新加载</button></div></div>
     <div v-else class="feature-grid"><article v-for="plugin in pluginStore.plugins" :key="plugin.plugin_id" class="item-card extension-card" @click="pluginStore.selectPlugin(plugin.plugin_id)"><div class="extension-title"><AppIcon :icon="Connection" :size="22" /><div><strong>{{ plugin.name }}</strong><p>v{{ plugin.version }}</p></div><span class="badge" :class="{ success: plugin.status === 'ready', error: plugin.status === 'error', warning: plugin.status === 'permission_required' }">{{ plugin.status }}</span></div><p class="muted">{{ plugin.description }}</p><p class="subtle">{{ plugin.permissions.length }} 项权限 · {{ plugin.contributions.length }} 项 Contribution</p></article></div>
   </section>
 </template>

frontend/src/features/settings/ModelRoutingSettings.spec.ts

@@ -0,0 +1,170 @@
+// @vitest-environment happy-dom
+import { flushPromises, mount } from '@vue/test-utils'
+import { afterEach, beforeEach, describe, expect, it, vi } from 'vitest'
+import type { ModelRoutingResponse, ProviderConfig } from '@/contracts'
+import { ApiErrorClass } from '@/services/apiClient'
+import * as service from '@/services/modelRoutingService'
+import { listProviders } from '@/services/providerService'
+import ModelRoutingSettings from './ModelRoutingSettings.vue'
+
+vi.mock('@/services/modelRoutingService', () => ({ getModelRouting: vi.fn(), saveModelRouting: vi.fn() }))
+vi.mock('@/services/providerService', () => ({ listProviders: vi.fn() }))
+const providers: ProviderConfig[] = [
+  { provider_id: 'p1', provider_type: 'openai_compatible', name: 'Custom API', enabled: true, default_model: 'chat-model', capabilities: {}, has_credential: true },
+  { provider_id: 'p2', provider_type: 'openai_chat', name: 'OpenAI', enabled: true, default_model: '', capabilities: {}, has_credential: true },
+  { provider_id: 'responses', provider_type: 'openai_responses', name: 'Responses', enabled: true, default_model: '', capabilities: {}, has_credential: true },
+  { provider_id: 'anthropic', provider_type: 'anthropic_messages', name: 'Anthropic', enabled: true, default_model: '', capabilities: {}, has_credential: true },
+  { provider_id: 'ollama', provider_type: 'ollama', name: 'Ollama', enabled: true, default_model: '', capabilities: {}, has_credential: false },
+  { provider_id: 'disabled', provider_type: 'openai_chat', name: 'Disabled', enabled: false, default_model: '', capabilities: {}, has_credential: true },
+]
+const initial: ModelRoutingResponse = { config: { version: 3, embedding: null, transcription: null, speaker_matching: null }, local_backends: [
+  { capability: 'embedding', status: 'placeholder', message: 'hash fallback' },
+  { capability: 'transcription', status: 'not_installed', message: 'ASR not installed' },
+  { capability: 'speaker_matching', status: 'not_installed', message: 'speaker not installed' },
+] }
+const wrappers: ReturnType<typeof mount>[] = []
+async function render() {
+  const wrapper = mount(ModelRoutingSettings)
+  wrappers.push(wrapper)
+  await flushPromises()
+  return wrapper
+}
+beforeEach(() => {
+  vi.resetAllMocks()
+  vi.mocked(service.getModelRouting).mockResolvedValue(structuredClone(initial))
+  vi.mocked(listProviders).mockResolvedValue(providers)
+  vi.mocked(service.saveModelRouting).mockImplementation(async config => ({ ...initial, config: { ...config, version: config.version + 1 } }))
+})
+afterEach(() => { wrappers.splice(0).forEach(wrapper => wrapper.unmount()) })
+
+describe('ModelRoutingSettings', () => {
+  it('loads local selections honestly, explains index rebuilds, and disables incompatible providers', async () => {
+    const wrapper = await render()
+    expect(wrapper.findAll('select').map(select => (select.element as HTMLSelectElement).value)).toEqual(['', '', ''])
+    expect(wrapper.text()).toContain('当前为占位实现')
+    expect(wrapper.text()).toContain('真实本地 ASR 尚未接入')
+    expect(wrapper.text()).toContain('真实本地说话人匹配尚未接入')
+    expect(wrapper.text()).toContain('重建全部')
+    expect(wrapper.text()).toContain('重建完成前继续使用本地检索')
+    expect(wrapper.text()).toContain('不是 OpenAI 标准接口')
+    for (const id of ['responses', 'anthropic', 'ollama', 'disabled']) expect(wrapper.get(`option[value="${id}"]`).attributes()).toHaveProperty('disabled')
+    expect(wrapper.get('option[value="p1"]').attributes()).not.toHaveProperty('disabled')
+  })
+
+  it('saves all three independent bindings with expected version and remote dimensions', async () => {
+    const wrapper = await render()
+    for (const capability of ['embedding', 'transcription', 'speaker_matching']) {
+      const card = wrapper.get(`[data-capability="${capability}"]`)
+      await card.get('select').setValue('p1')
+      await card.get('[data-field="model"]').setValue(`${capability}-model`)
+    }
+    await wrapper.get('[data-field="dimensions"]').setValue('3072')
+    await wrapper.get('form').trigger('submit')
+    await flushPromises()
+    expect(service.saveModelRouting).toHaveBeenCalledWith({ version: 3,
+      embedding: { provider_id: 'p1', model: 'embedding-model', endpoint: '/embeddings', dimensions: 3072 },
+      transcription: { provider_id: 'p1', model: 'transcription-model', endpoint: '/audio/transcriptions' },
+      speaker_matching: { provider_id: 'p1', model: 'speaker_matching-model', endpoint: '/audio/speaker-matches' },
+    })
+    expect(wrapper.text()).toContain('配置版本 4')
+    expect(wrapper.text()).toContain('模型路由已保存')
+    await wrapper.get('[data-capability="embedding"] select').setValue('')
+    await wrapper.get('form').trigger('submit')
+    await flushPromises()
+    expect(service.saveModelRouting).toHaveBeenLastCalledWith(expect.objectContaining({ version: 4, embedding: null }))
+  })
+
+  it('supports omitted dimensions and clears stale models/endpoints when switching providers', async () => {
+    const wrapper = await render()
+    const card = wrapper.get('[data-capability="embedding"]')
+    await card.get('select').setValue('p1')
+    await card.get('[data-field="model"]').setValue('custom-embedding')
+    await card.get('[data-field="endpoint"]').setValue('/custom/embeddings')
+    await wrapper.get('form').trigger('submit')
+    await flushPromises()
+    expect(service.saveModelRouting).toHaveBeenCalledWith(expect.objectContaining({ embedding: { provider_id: 'p1', model: 'custom-embedding', endpoint: '/custom/embeddings', dimensions: null } }))
+    await card.get('select').setValue('p2')
+    expect((card.get('[data-field="model"]').element as HTMLInputElement).value).toBe('')
+    expect((card.get('[data-field="endpoint"]').element as HTMLInputElement).value).toBe('/embeddings')
+  })
+
+  it('shows a loading state and does not offer a default local configuration after load failure', async () => {
+    let fail!: (error: Error) => void
+    vi.mocked(service.getModelRouting).mockReturnValueOnce(new Promise((_, reject) => { fail = reject }))
+    const wrapper = await render()
+    expect(wrapper.text()).toContain('正在加载模型路由')
+    expect(wrapper.find('form').exists()).toBe(false)
+    fail(new Error('offline'))
+    await flushPromises()
+    expect(wrapper.text()).toContain('加载失败：offline')
+    expect(wrapper.find('form').exists()).toBe(false)
+    await wrapper.get('button').trigger('click')
+    await flushPromises()
+    expect(wrapper.find('form').exists()).toBe(true)
+  })
+
+  it('keeps unsaved input on save failure and retries without inventing a new version', async () => {
+    const wrapper = await render()
+    const card = wrapper.get('[data-capability="transcription"]')
+    await card.get('select').setValue('p1')
+    await card.get('[data-field="model"]').setValue('asr-model')
+    vi.mocked(service.saveModelRouting).mockRejectedValueOnce(new Error('disk full'))
+    await wrapper.get('form').trigger('submit')
+    await flushPromises()
+    expect(wrapper.text()).toContain('保存失败：disk full')
+    expect((card.get('[data-field="model"]').element as HTMLInputElement).value).toBe('asr-model')
+    await wrapper.get('form').trigger('submit')
+    await flushPromises()
+    expect(service.saveModelRouting).toHaveBeenLastCalledWith(expect.objectContaining({ version: 3 }))
+  })
+
+  it('blocks overwrite after a conflict until explicitly reloading the latest configuration', async () => {
+    const wrapper = await render()
+    vi.mocked(service.saveModelRouting).mockRejectedValueOnce(new ApiErrorClass('MODEL_ROUTING_VERSION_CONFLICT', 'stale'))
+    await wrapper.get('form').trigger('submit')
+    await flushPromises()
+    expect(wrapper.text()).toContain('配置版本冲突')
+    expect(wrapper.get('button[type="submit"]').attributes()).toHaveProperty('disabled')
+    await wrapper.get('form').trigger('submit')
+    expect(service.saveModelRouting).toHaveBeenCalledTimes(1)
+    vi.mocked(service.getModelRouting).mockResolvedValueOnce({ ...initial, config: { ...initial.config, version: 8 } })
+    await wrapper.findAll('button').find(button => button.text().includes('放弃当前输入'))!.trigger('click')
+    await flushPromises()
+    await wrapper.get('form').trigger('submit')
+    await flushPromises()
+    expect(service.saveModelRouting).toHaveBeenLastCalledWith(expect.objectContaining({ version: 8 }))
+  })
+
+  it('prevents invalid dimensions, endpoints, and missing provider bindings from being saved', async () => {
+    vi.mocked(service.getModelRouting).mockResolvedValueOnce({ ...initial, config: { ...initial.config, embedding: { provider_id: 'missing', model: 'old-model', endpoint: '/embeddings' } } })
+    const wrapper = await render()
+    expect(wrapper.text()).toContain('原提供商已不可用')
+    await wrapper.get('form').trigger('submit')
+    expect(service.saveModelRouting).not.toHaveBeenCalled()
+    const card = wrapper.get('[data-capability="embedding"]')
+    await card.get('select').setValue('p1')
+    await card.get('[data-field="model"]').setValue('embedding-model')
+    await card.get('[data-field="dimensions"]').setValue('1.5')
+    await wrapper.get('form').trigger('submit')
+    expect(wrapper.text()).toContain('嵌入维度必须为 1–16384 的整数')
+    await card.get('[data-field="dimensions"]').setValue('16385')
+    await wrapper.get('form').trigger('submit')
+    expect(service.saveModelRouting).not.toHaveBeenCalled()
+    expect(card.get('[data-field="dimensions"]').attributes('max')).toBe('16384')
+    await card.get('[data-field="dimensions"]').setValue('')
+    await card.get('[data-field="endpoint"]').setValue('https://example.test/embeddings')
+    await wrapper.get('form').trigger('submit')
+    expect(wrapper.text()).toContain('Endpoint 必须是以 / 开头的相对路径')
+    expect(service.saveModelRouting).not.toHaveBeenCalled()
+  })
+
+  it('labels injected ready local backends accurately', async () => {
+    vi.mocked(service.getModelRouting).mockResolvedValueOnce({ ...initial, local_backends: [{ capability: 'transcription', status: 'ready', message: 'Local ASR ready' }] })
+    const wrapper = await render()
+    const card = wrapper.get('[data-capability="transcription"]')
+    expect(card.get('option[value=""]').text()).toBe('本地 · 已就绪')
+    expect(card.text()).toContain('本地后端已就绪')
+    expect(card.text()).toContain('Local ASR ready')
+    expect(card.text()).not.toContain('真实本地 ASR 尚未接入')
+  })
+})

frontend/src/features/settings/ModelRoutingSettings.vue

@@ -0,0 +1,160 @@
+<script setup lang="ts">
+import { computed, onBeforeUnmount, onMounted, reactive, ref } from 'vue'
+import type { ModelBinding, ModelRoutingConfig, ModelRoutingResponse, ProviderConfig, RoutingCapability } from '@/contracts'
+import { getModelRouting, saveModelRouting } from '@/services/modelRoutingService'
+import { listProviders } from '@/services/providerService'
+import { ApiErrorClass } from '@/services/apiClient'
+
+const capabilities: Array<{ id: RoutingCapability; name: string; endpoint: string; placeholder: string; local: string }> = [
+  { id: 'embedding', name: '向量嵌入 · Embedding', endpoint: '/embeddings', placeholder: '例如 text-embedding-3-small', local: '当前为占位实现，尚未接入真实本地嵌入模型。' },
+  { id: 'transcription', name: '语音转写 · Transcription', endpoint: '/audio/transcriptions', placeholder: '输入转写模型 ID', local: '真实本地 ASR 尚未接入，等待阶段 F；当前无法进行本地语音识别。' },
+  { id: 'speaker_matching', name: '说话人匹配 · Speaker matching', endpoint: '/audio/speaker-matches', placeholder: '输入说话人匹配模型 ID', local: '真实本地说话人匹配尚未接入，等待阶段 F；当前无法进行本地声纹匹配。' },
+]
+type Draft = { provider_id: string; model: string; endpoint: string; dimensions: string | number }
+const drafts = reactive(Object.fromEntries(capabilities.map(item => [item.id, { provider_id: '', model: '', endpoint: item.endpoint, dimensions: '' }])) as Record<RoutingCapability, Draft>)
+const providers = ref<ProviderConfig[]>([])
+const response = ref<ModelRoutingResponse | null>(null)
+const loading = ref(false)
+const saving = ref(false)
+const error = ref('')
+const saved = ref(false)
+const conflict = ref(false)
+let active = true
+const eligible = (provider: ProviderConfig) => provider.enabled && ['openai_chat', 'openai_compatible'].includes(provider.provider_type)
+const available = computed(() => providers.value.filter(eligible))
+const unavailable = computed(() => providers.value.filter(provider => !eligible(provider)))
+const localBackend = (capability: RoutingCapability) => response.value?.local_backends.find(item => item.capability === capability)
+const localLabel = (capability: RoutingCapability) => {
+  const status = localBackend(capability)?.status
+  return status === 'ready' ? '已就绪' : status === 'placeholder' ? '占位实现' : '尚未接入'
+}
+const protocols = [
+  { id: 'openai_chat', label: 'OpenAI Chat' }, { id: 'openai_compatible', label: 'OpenAI Compatible' },
+  { id: 'openai_responses', label: 'Responses' }, { id: 'anthropic_messages', label: 'Anthropic' }, { id: 'ollama', label: 'Ollama' },
+]
+
+function applyResponse(result: ModelRoutingResponse) {
+  response.value = result
+  for (const item of capabilities) {
+    const binding = result.config[item.id]
+    Object.assign(drafts[item.id], { provider_id: binding?.provider_id ?? '', model: binding?.model ?? '', endpoint: binding?.endpoint ?? item.endpoint, dimensions: binding?.dimensions?.toString() ?? '' })
+  }
+}
+
+async function load() {
+  if (loading.value || saving.value) return
+  loading.value = true
+  error.value = ''
+  saved.value = false
+  try {
+    const [routing, items] = await Promise.all([getModelRouting(), listProviders()])
+    if (!active) return
+    providers.value = items
+    applyResponse(routing)
+    conflict.value = false
+  } catch (reason) {
+    if (active) error.value = `加载失败：${reason instanceof Error ? reason.message : '无法读取模型路由或提供商'}`
+  } finally { loading.value = false }
+}
+
+onMounted(load)
+onBeforeUnmount(() => { active = false })
+
+function changeProvider(capability: RoutingCapability) {
+  const draft = drafts[capability]
+  draft.model = ''
+  draft.dimensions = ''
+  draft.endpoint = capabilities.find(item => item.id === capability)!.endpoint
+  saved.value = false
+}
+
+function bindingFor(capability: RoutingCapability): ModelBinding | null {
+  const draft = drafts[capability]
+  if (!draft.provider_id) return null
+  if (!available.value.some(provider => provider.provider_id === draft.provider_id)) throw new Error('请选择已启用且协议可用的提供商，或切换到本地。')
+  if (!draft.model.trim()) throw new Error('请填写所选 API 的模型 ID。')
+  if (!/^\/[A-Za-z0-9_/-]+$/.test(draft.endpoint) || draft.endpoint.startsWith('//')) throw new Error('Endpoint 必须是以 / 开头的相对路径，只能包含字母、数字、下划线、连字符和 /。')
+  const binding: ModelBinding = { provider_id: draft.provider_id, model: draft.model.trim(), endpoint: draft.endpoint }
+  if (capability === 'embedding') {
+    const dimension = String(draft.dimensions).trim()
+    if (dimension && (!/^\d+$/.test(dimension) || !Number.isSafeInteger(Number(dimension)) || Number(dimension) < 1 || Number(dimension) > 16384)) throw new Error('嵌入维度必须为 1–16384 的整数，或留空使用 API 默认值。')
+    binding.dimensions = dimension ? Number(dimension) : null
+  }
+  return binding
+}
+
+async function save() {
+  if (!response.value || loading.value || saving.value || conflict.value) return
+  saving.value = true
+  error.value = ''
+  saved.value = false
+  try {
+    const config: ModelRoutingConfig = {
+      version: response.value.config.version,
+      embedding: bindingFor('embedding'), transcription: bindingFor('transcription'), speaker_matching: bindingFor('speaker_matching'),
+    }
+    const result = await saveModelRouting(config)
+    if (active) { applyResponse(result); saved.value = true }
+  } catch (reason) {
+    if (!active) return
+    conflict.value = reason instanceof ApiErrorClass && /CONFLICT|VERSION|HTTP_409/i.test(reason.code)
+    error.value = conflict.value
+      ? '配置版本冲突：其他窗口已修改路由。当前输入尚未保存，请重新加载最新配置后再编辑。'
+      : `保存失败：${reason instanceof Error ? reason.message : '请重试'}`
+  } finally { saving.value = false }
+}
+</script>
+
+<template>
+  <section class="routing-settings" aria-labelledby="routing-title" :aria-busy="loading || saving">
+    <div><h2 id="routing-title">能力模型路由</h2><p class="subtle">向量嵌入、语音转写和说话人匹配分别选择提供商与模型，独立于默认聊天模型。API Key 在「模型提供商」中管理。</p></div>
+    <p class="subtle">未选择提供商即使用本地路径。API 请求失败、配置不可用或响应无效时，服务端会回退到当前本地处理；本地占位不代表真实模型已接入。</p>
+    <p v-if="loading" role="status">正在加载模型路由…</p>
+    <div v-if="error" class="error-banner" role="alert">{{ error }}</div>
+    <div class="inline-actions"><button type="button" class="button-secondary" :disabled="loading || saving" @click="load">{{ conflict ? '放弃当前输入并加载最新配置' : response ? '重新加载（放弃未保存更改）' : '重试加载' }}</button><span v-if="response" class="subtle">配置版本 {{ response.config.version }}</span></div>
+    <form v-if="response" @submit.prevent="save" @input="saved = false" @change="saved = false">
+      <fieldset :disabled="loading || saving || conflict">
+        <article v-for="capability in capabilities" :key="capability.id" class="routing-card" :data-capability="capability.id">
+          <h3>{{ capability.name }}</h3>
+          <p v-if="capability.id === 'embedding'" class="embedding-notice">更换模型或接口后，请重建全部索引。重建完成前继续使用本地检索。</p>
+          <div class="protocols" aria-label="协议可用性">
+            <span v-for="protocol in protocols" :key="protocol.id" class="badge" :class="{ 'protocol-unavailable': !['openai_chat', 'openai_compatible'].includes(protocol.id) }">{{ protocol.label }}{{ ['openai_chat', 'openai_compatible'].includes(protocol.id) ? ' · 可用' : ' · 不可用' }}</span>
+          </div>
+          <label class="field"><span>处理方式 / 提供商</span><select v-model="drafts[capability.id].provider_id" class="select" data-field="provider" @change="changeProvider(capability.id)">
+            <option value="">本地 · {{ localLabel(capability.id) }}</option>
+            <option v-for="provider in available" :key="provider.provider_id" :value="provider.provider_id">{{ provider.name }} · {{ provider.provider_type }}</option>
+            <option v-for="provider in unavailable" :key="provider.provider_id" :value="provider.provider_id" disabled>{{ provider.name }} · {{ provider.enabled ? '协议不可用' : '未启用' }}</option>
+            <option v-if="drafts[capability.id].provider_id && !providers.some(provider => provider.provider_id === drafts[capability.id].provider_id)" :value="drafts[capability.id].provider_id" disabled>原提供商已不可用 · {{ drafts[capability.id].provider_id }}</option>
+          </select></label>
+          <div v-if="drafts[capability.id].provider_id" class="routing-fields">
+            <label class="field"><span>模型 ID</span><input v-model="drafts[capability.id].model" class="input" data-field="model" :placeholder="capability.placeholder" maxlength="256" required /></label>
+            <label class="field"><span>Endpoint（相对 Base URL）</span><input v-model="drafts[capability.id].endpoint" class="input" data-field="endpoint" :placeholder="capability.endpoint" maxlength="256" required /></label>
+            <label v-if="capability.id === 'embedding'" class="field"><span>向量维度（可选）</span><input v-model="drafts.embedding.dimensions" class="input" data-field="dimensions" type="number" min="1" max="16384" step="1" placeholder="留空使用 API 默认维度" /><small class="subtle">填写模型支持的 1–16384 整数维度，或留空使用 API 默认值。</small></label>
+          </div>
+          <p v-if="capability.id === 'speaker_matching'" class="subtle">说话人匹配使用本应用自定义 HTTP multipart 契约。该端点不是 OpenAI 标准接口；服务需实现对应的说话人匹配请求和响应。</p>
+          <div class="local-status" :class="{ selected: !drafts[capability.id].provider_id }">
+            <strong>{{ drafts[capability.id].provider_id ? '本地回退状态' : '当前本地状态' }}</strong>
+            <p>{{ localBackend(capability.id)?.status === 'ready' ? '本地后端已就绪。' : capability.local }}</p>
+            <p v-for="backend in response.local_backends.filter(item => item.capability === capability.id)" :key="backend.capability" class="subtle"><span class="badge">{{ backend.status === 'ready' ? '已就绪' : backend.status === 'placeholder' ? '占位实现' : '未安装 / 未接入' }}</span> {{ backend.message }}</p>
+          </div>
+        </article>
+      </fieldset>
+      <div class="inline-actions"><button type="submit" class="button-primary" :disabled="loading || saving || conflict">{{ saving ? '保存中…' : '保存模型路由' }}</button><span v-if="saved" role="status">模型路由已保存。</span></div>
+    </form>
+  </section>
+</template>
+
+<style scoped>
+.routing-settings, form, fieldset { display: grid; gap: var(--space-lg); }
+.routing-settings { border-top: 1px solid var(--color-border-default); padding-top: var(--space-xl); margin-top: var(--space-md); }
+fieldset { min-width: 0; padding: 0; margin: 0; border: 0; }
+.routing-card { display: grid; gap: var(--space-md); padding: var(--space-lg); border: 1px solid var(--color-border-default); border-radius: var(--radius-lg); background: var(--color-surface-primary); }
+.routing-card h3 { margin: 0; }
+.protocols { display: flex; flex-wrap: wrap; gap: var(--space-xs); }
+.protocol-unavailable { opacity: .65; }
+.embedding-notice { padding: var(--space-md); border-radius: var(--radius-md); background: var(--color-accent-soft); color: var(--color-text-primary); }
+.routing-fields { display: grid; grid-template-columns: repeat(2, minmax(0, 1fr)); gap: var(--space-md); }
+.local-status { display: grid; gap: var(--space-xs); padding: var(--space-md); background: var(--color-background-secondary); border-radius: var(--radius-md); }
+.local-status.selected { border-left: 3px solid var(--color-accent-primary); }
+@media (max-width: 700px) { .routing-fields { grid-template-columns: 1fr; } }
+</style>

frontend/src/features/settings/ProviderForm.spec.ts

@@ -0,0 +1,157 @@
+// @vitest-environment happy-dom
+import { flushPromises, mount } from '@vue/test-utils'
+import { afterEach, beforeEach, describe, expect, it, vi } from 'vitest'
+import type { ProviderConfig, ProviderPreset } from '@/contracts'
+import * as service from '@/services/providerService'
+import ProviderForm from './ProviderForm.vue'
+import ProviderPresetSelector from './ProviderPresetSelector.vue'
+
+vi.mock('@/services/providerService', () => ({ listProviderPresets: vi.fn(), getCredentialStatus: vi.fn(), putCredential: vi.fn(), createProvider: vi.fn(), updateProvider: vi.fn() }))
+const presets: ProviderPreset[] = [
+  { preset_id: 'deepseek', name: 'DeepSeek', provider_type: 'openai_compatible', base_url: 'https://deepseek.example.test', default_credential_id: 'shared-deepseek', requires_credential: true, logo_id: 'deepseek' },
+  { preset_id: 'qwen', name: '通义千问', provider_type: 'openai_compatible', base_url: 'https://qwen.example.test', default_credential_id: 'shared-qwen', requires_credential: true, logo_id: 'qwen' },
+]
+const existing: ProviderConfig = { provider_id: 'p1', provider_type: 'openai_compatible', name: 'DeepSeek', base_url: presets[0].base_url, default_model: 'old-model', enabled: true, credential_id: 'old-shared-key', has_credential: true, capabilities: {} }
+const wrappers: ReturnType<typeof mount>[] = []
+async function render(provider?: ProviderConfig) {
+  const wrapper = mount(ProviderForm, { props: { provider, models: [{ model_id: 'old-model', name: 'Old', capabilities: {} }] } })
+  wrappers.push(wrapper)
+  await flushPromises()
+  return wrapper
+}
+beforeEach(() => {
+  vi.resetAllMocks()
+  vi.mocked(service.listProviderPresets).mockResolvedValue(presets)
+  vi.mocked(service.getCredentialStatus).mockResolvedValue(true)
+  vi.mocked(service.putCredential).mockResolvedValue()
+  vi.mocked(service.createProvider).mockResolvedValue(existing)
+  vi.mocked(service.updateProvider).mockResolvedValue(existing)
+})
+afterEach(() => { wrappers.splice(0).forEach(wrapper => wrapper.unmount()) })
+
+describe('ProviderForm', () => {
+  it('filters compact preset chips and resolves bundled logos', async () => {
+    const wrapper = await render()
+    await wrapper.get('#provider-search').setValue('通义')
+    expect(wrapper.find('[data-preset="deepseek"]').exists()).toBe(false)
+    expect(wrapper.find('[data-preset="qwen"]').exists()).toBe(true)
+    expect(wrapper.get('[data-preset="qwen"] img').attributes('src')).not.toMatch(/^https?:/)
+  })
+
+  it('clears the secret, old model and credential on preset and custom selection', async () => {
+    const wrapper = await render(existing)
+    await wrapper.get('input[type="password"]').setValue('draft-secret')
+    await wrapper.get('[data-preset="qwen"]').trigger('click')
+    expect((wrapper.get('input[type="password"]').element as HTMLInputElement).value).toBe('')
+    expect((wrapper.get('[data-field="model"]').element as HTMLInputElement).value).toBe('')
+    expect(wrapper.findAll('datalist option')).toHaveLength(0)
+    await wrapper.get('form').trigger('submit')
+    expect(service.updateProvider).not.toHaveBeenCalled()
+    expect(wrapper.text()).toContain('请输入 API Key')
+    await wrapper.get('input[type="password"]').setValue('new-secret')
+    wrapper.getComponent(ProviderPresetSelector).vm.$emit('update:modelValue', '')
+    await flushPromises()
+    expect((wrapper.get('input[type="password"]').element as HTMLInputElement).value).toBe('')
+  })
+
+  it('allocates different credential IDs for two new providers using the same preset', async () => {
+    for (let i = 0; i < 2; i++) {
+      const wrapper = await render()
+      await wrapper.get('[data-preset="deepseek"]').trigger('click')
+      await wrapper.get('input[type="password"]').setValue(`test-key-${i}`)
+      await wrapper.get('form').trigger('submit')
+      await flushPromises()
+    }
+    const ids = vi.mocked(service.putCredential).mock.calls.map(call => call[0])
+    expect(ids).toHaveLength(2)
+    expect(new Set(ids).size).toBe(2)
+    ids.forEach(id => expect(id).toMatch(/^provider-key-[0-9a-f-]{36}$/))
+    vi.mocked(service.createProvider).mock.calls.forEach(([data], index) => {
+      expect(data.credential_id).toBe(ids[index])
+      expect(JSON.stringify(data)).not.toContain('test-key')
+      expect(JSON.stringify(data)).not.toContain('shared-deepseek')
+    })
+  })
+
+  it('accepts a custom API key and clears it after a failed credential save', async () => {
+    const wrapper = await render()
+    await wrapper.get('[data-field="name"]').setValue('Custom')
+    await wrapper.get('[data-field="base-url"]').setValue('https://custom.example.test/v1')
+    await wrapper.get('input[type="password"]').setValue('custom-test-key')
+    vi.mocked(service.putCredential).mockRejectedValueOnce(new Error('credential store unavailable'))
+    await wrapper.get('form').trigger('submit')
+    await flushPromises()
+    expect((wrapper.get('input[type="password"]').element as HTMLInputElement).value).toBe('')
+    expect(wrapper.text()).toContain('credential store unavailable')
+    expect(service.createProvider).not.toHaveBeenCalled()
+    await wrapper.get('input[type="password"]').setValue('custom-test-key')
+    await wrapper.get('form').trigger('submit')
+    await flushPromises()
+    expect(service.createProvider).toHaveBeenCalledWith(expect.objectContaining({ name: 'Custom', credential_id: expect.stringMatching(/^provider-key-/) }))
+  })
+
+  it('preserves its own existing key when untouched and rotates shared legacy references when replacing a key', async () => {
+    const untouched = await render(existing)
+    await untouched.get('form').trigger('submit')
+    await flushPromises()
+    expect(service.updateProvider).toHaveBeenLastCalledWith('p1', expect.objectContaining({ credential_id: 'old-shared-key', default_model: 'old-model' }))
+    const rotated = await render(existing)
+    await rotated.get('input[type="password"]').setValue('replacement-test-key')
+    await rotated.get('form').trigger('submit')
+    await flushPromises()
+    expect(service.putCredential).toHaveBeenCalledWith(expect.stringMatching(/^provider-key-/), 'replacement-test-key')
+    expect(service.updateProvider).toHaveBeenLastCalledWith('p1', expect.objectContaining({ credential_id: vi.mocked(service.putCredential).mock.calls[0][0] }))
+  })
+
+  it('persists edited protocols and unlinks the previous credential and model', async () => {
+    const wrapper = await render(existing)
+    await wrapper.get('[data-field="protocol"]').setValue('openai_responses')
+    await wrapper.get('form').trigger('submit')
+    await flushPromises()
+    expect(service.updateProvider).toHaveBeenCalledWith('p1', expect.objectContaining({ provider_type: 'openai_responses', default_model: '', credential_id: null }))
+  })
+
+  it('does not let a late credential status reuse a key after switching presets', async () => {
+    let resolveStatus!: (configured: boolean) => void
+    vi.mocked(service.getCredentialStatus).mockReturnValue(new Promise(resolve => { resolveStatus = resolve }))
+    const wrapper = await render(existing)
+    await wrapper.get('[data-preset="qwen"]').trigger('click')
+    resolveStatus(true)
+    await flushPromises()
+    await wrapper.get('form').trigger('submit')
+    expect(service.updateProvider).not.toHaveBeenCalled()
+    expect(wrapper.text()).toContain('请输入 API Key')
+  })
+
+  it('clears secrets on close and unmount, and stops a pending credential save from creating a provider', async () => {
+    let finish!: () => void
+    vi.mocked(service.putCredential).mockReturnValue(new Promise(resolve => { finish = resolve }))
+    const wrapper = await render()
+    await wrapper.get('[data-preset="deepseek"]').trigger('click')
+    await wrapper.get('input[type="password"]').setValue('pending-test-key')
+    await wrapper.get('form').trigger('submit')
+    await wrapper.get('[aria-label="关闭提供商表单"]').trigger('click')
+    expect((wrapper.get('input[type="password"]').element as HTMLInputElement).value).toBe('')
+    wrapper.unmount()
+    finish()
+    await flushPromises()
+    expect(service.createProvider).not.toHaveBeenCalled()
+    const reopened = await render()
+    expect((reopened.get('input[type="password"]').element as HTMLInputElement).value).toBe('')
+  })
+
+  it('clears a secret after provider save failure and keeps only the successfully saved reference for retry', async () => {
+    const wrapper = await render()
+    await wrapper.get('[data-preset="deepseek"]').trigger('click')
+    await wrapper.get('input[type="password"]').setValue('retry-test-key')
+    vi.mocked(service.createProvider).mockRejectedValueOnce(new Error('provider save failed'))
+    await wrapper.get('form').trigger('submit')
+    await flushPromises()
+    expect(wrapper.text()).toContain('provider save failed')
+    expect((wrapper.get('input[type="password"]').element as HTMLInputElement).value).toBe('')
+    await wrapper.get('form').trigger('submit')
+    await flushPromises()
+    expect(service.putCredential).toHaveBeenCalledTimes(1)
+    expect(service.createProvider).toHaveBeenCalledTimes(2)
+  })
+})

frontend/src/features/settings/ProviderForm.vue

@@ -0,0 +1,177 @@
+<script setup lang="ts">
+import { computed, nextTick, onBeforeUnmount, onMounted, reactive, ref } from 'vue'
+import type { ModelInfo, ProviderConfig, ProviderPreset, ProviderType } from '@/contracts'
+import * as service from '@/services/providerService'
+import ProviderPresetSelector from './ProviderPresetSelector.vue'
+
+const props = defineProps<{ provider?: ProviderConfig; models?: ModelInfo[] }>()
+const emit = defineEmits<{ close: []; saved: [provider: ProviderConfig] }>()
+const newCredentialId = () => `provider-key-${crypto.randomUUID()}`
+const form = reactive({
+  preset_id: '', provider_type: props.provider?.provider_type ?? 'openai_compatible' as ProviderType,
+  name: props.provider?.name ?? '', base_url: props.provider?.base_url ?? '',
+  default_model: props.provider?.default_model ?? '', enabled: props.provider?.enabled ?? true,
+})
+const credentialId = ref(props.provider?.credential_id || newCredentialId())
+const apiKey = ref('')
+const configured = ref(false)
+const credentialLoading = ref(false)
+const credentialError = ref('')
+const presets = ref<ProviderPreset[]>([])
+const presetsLoading = ref(false)
+const presetsError = ref('')
+const saving = ref(false)
+const error = ref('')
+const contextChanged = ref(false)
+const dialog = ref<HTMLElement>()
+const previousFocus = document.activeElement as HTMLElement | null
+let active = true
+let credentialGeneration = 0
+const selectedPreset = computed(() => presets.value.find(preset => preset.preset_id === form.preset_id))
+const modelOptions = computed(() => contextChanged.value ? [] : props.models ?? [])
+
+async function loadPresets() {
+  presetsLoading.value = true
+  presetsError.value = ''
+  try {
+    presets.value = await service.listProviderPresets()
+    if (!contextChanged.value) form.preset_id = presets.value.find(preset => preset.provider_type === props.provider?.provider_type && preset.base_url === props.provider?.base_url)?.preset_id ?? ''
+  } catch { presetsError.value = '预设加载失败，请重试，或填写自定义服务。' }
+  finally { presetsLoading.value = false }
+}
+
+onMounted(async () => {
+  void loadPresets()
+  if (props.provider?.credential_id) {
+    const generation = credentialGeneration
+    credentialLoading.value = true
+    try {
+      const result = await service.getCredentialStatus(credentialId.value)
+      if (active && generation === credentialGeneration) configured.value = result
+    } catch {
+      if (active && generation === credentialGeneration) credentialError.value = '无法检查已保存的凭据。可输入新密钥，或关闭后重试。'
+    } finally {
+      if (generation === credentialGeneration) credentialLoading.value = false
+    }
+  }
+  await nextTick()
+  if (active) dialog.value?.querySelector<HTMLInputElement>('input')?.focus()
+})
+
+function detachCredential() {
+  credentialGeneration++
+  apiKey.value = ''
+  credentialId.value = newCredentialId()
+  configured.value = false
+  credentialLoading.value = false
+  credentialError.value = ''
+  form.default_model = ''
+  contextChanged.value = true
+  error.value = ''
+}
+
+function applyPreset(id: string) {
+  if (form.preset_id === id) return
+  detachCredential()
+  form.preset_id = id
+  const preset = presets.value.find(item => item.preset_id === id)
+  Object.assign(form, { provider_type: preset?.provider_type ?? 'openai_compatible', name: preset?.name ?? '', base_url: preset?.base_url ?? '' })
+}
+
+function changeConnection() {
+  form.preset_id = ''
+  detachCredential()
+}
+
+function close() {
+  active = false
+  apiKey.value = ''
+  emit('close')
+}
+
+onBeforeUnmount(() => {
+  active = false
+  apiKey.value = ''
+  previousFocus?.focus()
+})
+
+function handleKeydown(event: KeyboardEvent) {
+  if (event.key === 'Escape') { event.preventDefault(); close() }
+  if (event.key !== 'Tab') return
+  const elements = Array.from(dialog.value?.querySelectorAll<HTMLElement>('button, input, select, [tabindex="0"]') ?? []).filter(element => !element.matches(':disabled'))
+  const first = elements[0], last = elements[elements.length - 1]
+  if (event.shiftKey && document.activeElement === first) { event.preventDefault(); last?.focus() }
+  else if (!event.shiftKey && document.activeElement === last) { event.preventDefault(); first?.focus() }
+}
+
+async function save() {
+  if (saving.value || credentialLoading.value || !active) return
+  error.value = ''
+  saving.value = true
+  try {
+    if (!form.name.trim() || !form.base_url.trim()) throw new Error('请填写名称和 Base URL。')
+    if (selectedPreset.value?.requires_credential && !apiKey.value.trim() && !configured.value) throw new Error('请输入 API Key。密钥将由后端加密保存。')
+    // Snapshot before awaiting: closing/unmounting must never create a provider with a changed draft.
+    const data = { provider_type: form.provider_type, name: form.name.trim(), base_url: form.base_url.trim() || undefined, default_model: form.default_model.trim(), enabled: form.enabled, capabilities: {}, has_credential: false }
+    if (apiKey.value.trim()) {
+      // Rotate even an existing reference: older installations may share preset credential IDs.
+      const nextId = newCredentialId()
+      const request = service.putCredential(nextId, apiKey.value.trim())
+      apiKey.value = ''
+      await request
+      if (!active) return
+      credentialId.value = nextId
+      configured.value = true
+    }
+    const reference = configured.value ? credentialId.value : undefined
+    // A failed status check must not silently unlink the provider's existing credential.
+    if (credentialError.value && !reference) throw new Error(credentialError.value)
+    const saved = props.provider
+      ? await service.updateProvider(props.provider.provider_id, { ...data, credential_id: reference ?? null })
+      : await service.createProvider({ ...data, credential_id: reference })
+    if (active) { emit('saved', saved); close() }
+  } catch (reason) {
+    if (active) error.value = reason instanceof Error ? reason.message : 'Provider 保存失败，请重试。'
+  } finally { apiKey.value = ''; saving.value = false }
+}
+</script>
+
+<template>
+  <div class="modal-backdrop provider-backdrop" @click.self="close" @keydown="handleKeydown">
+    <div ref="dialog" class="modal provider-modal" role="dialog" aria-modal="true" aria-labelledby="provider-form-title" :aria-busy="saving">
+      <div class="form-heading"><h2 id="provider-form-title">{{ provider ? '编辑 Provider' : '新增 Provider' }}</h2><button type="button" class="button-secondary" aria-label="关闭提供商表单" @click="close">关闭</button></div>
+      <p v-if="presetsLoading" class="subtle" role="status">正在加载提供商预设…</p>
+      <div v-if="presetsError" class="error-banner" role="alert">{{ presetsError }} <button type="button" class="button-secondary" :disabled="presetsLoading || saving" @click="loadPresets">重试</button></div>
+      <form @submit.prevent="save">
+        <fieldset :disabled="saving">
+          <ProviderPresetSelector :presets="presets" :model-value="form.preset_id" @update:model-value="applyPreset" />
+          <p v-if="selectedPreset?.description" class="subtle">{{ selectedPreset.description }}</p>
+          <div class="form-grid">
+            <label class="field"><span>接入协议</span><select v-model="form.provider_type" class="select" data-field="protocol" @change="changeConnection"><option value="openai_compatible">OpenAI Compatible</option><option value="openai_chat">OpenAI Chat</option><option value="openai_responses">OpenAI Responses</option><option value="anthropic_messages">Anthropic Messages</option><option value="ollama">Ollama</option></select></label>
+            <label class="field"><span>名称</span><input v-model="form.name" class="input" data-field="name" required /></label>
+            <label class="field wide"><span>Base URL</span><input v-model="form.base_url" class="input" data-field="base-url" placeholder="https://api.example.com/v1" required @change="changeConnection" /></label>
+            <label class="field wide"><span>API Key</span><input v-model="apiKey" class="input" type="password" autocomplete="new-password" spellcheck="false" :placeholder="configured ? '已配置，留空表示不修改' : '请输入 API Key（无鉴权服务可留空）'" /><small class="subtle">密钥由本地 AI Core 加密保存；提供商配置仅保存独立的凭据引用。</small></label>
+            <p v-if="credentialLoading" class="subtle wide" role="status">正在检查凭据状态…</p>
+            <p v-if="credentialError" class="error-text wide" role="alert">{{ credentialError }}</p>
+            <label class="field wide"><span>默认聊天模型</span><input v-model="form.default_model" class="input" data-field="model" list="provider-model-options" placeholder="输入模型 ID，或保存后获取模型列表" /><datalist id="provider-model-options"><option v-for="model in modelOptions" :key="model.model_id" :value="model.model_id">{{ model.name }}</option></datalist></label>
+          </div>
+          <label class="inline-actions"><input v-model="form.enabled" type="checkbox" /> 启用</label>
+        </fieldset>
+        <div v-if="error" class="error-banner" role="alert">{{ error }}</div>
+        <div class="inline-actions form-footer"><button class="button-primary" type="submit" :disabled="saving || credentialLoading">{{ saving ? '保存中…' : '保存提供商' }}</button><button type="button" class="button-secondary" @click="close">取消</button></div>
+      </form>
+    </div>
+  </div>
+</template>
+
+<style scoped>
+.provider-modal { width: min(820px, 100%); max-height: 90dvh; }
+.form-heading { display: flex; align-items: center; justify-content: space-between; gap: var(--space-md); margin-bottom: var(--space-md); }
+.form-heading h2 { margin: 0; }
+fieldset { display: grid; gap: var(--space-md); border: 0; padding: 0; margin: 0; min-width: 0; }
+.form-grid { display: grid; grid-template-columns: 1fr 1fr; gap: var(--space-md); }
+.wide { grid-column: 1 / -1; }
+.error-text { color: var(--color-error); }
+.form-footer { padding-top: var(--space-sm); }
+@media (max-width: 600px) { .provider-backdrop { padding: 12px; }.provider-modal { padding: var(--space-lg); max-height: 94dvh; }.form-grid { grid-template-columns: 1fr; } }
+</style>

frontend/src/features/settings/ProviderLogo.vue

@@ -0,0 +1,24 @@
+<script setup lang="ts">
+import { computed } from 'vue'
+
+const props = defineProps<{ logoId?: string }>()
+const assets = import.meta.glob<string>('../../assets/providers/*.svg', { eager: true, query: '?url', import: 'default' })
+const source = computed(() => {
+  const id = props.logoId === 'openai-responses' ? 'openai' : props.logoId
+  return assets[`../../assets/providers/${id}.svg`]
+})
+</script>
+
+<template>
+  <span class="provider-logo" :class="{ 'dark-logo': logoId === 'kimi' }" aria-hidden="true">
+    <img v-if="source" :src="source" alt="" width="22" height="22" />
+    <span v-else class="custom-logo">+</span>
+  </span>
+</template>
+
+<style scoped>
+.provider-logo { display: inline-flex; flex: 0 0 28px; align-items: center; justify-content: center; width: 28px; height: 28px; border-radius: 7px; background: #fff; color: #252b36; }
+img { display: block; object-fit: contain; }
+.dark-logo { background: #111; }
+.custom-logo { font-size: 23px; line-height: 1; }
+</style>

frontend/src/features/settings/ProviderPresetSelector.vue

@@ -0,0 +1,36 @@
+<script setup lang="ts">
+import { computed, ref } from 'vue'
+import type { ProviderPreset } from '@/contracts'
+import ProviderLogo from './ProviderLogo.vue'
+
+const props = defineProps<{ presets: ProviderPreset[]; modelValue: string }>()
+const emit = defineEmits<{ 'update:modelValue': [value: string] }>()
+const search = ref('')
+const filtered = computed(() => {
+  const query = search.value.trim().toLocaleLowerCase()
+  return props.presets.filter(preset => [preset.name, preset.preset_id, preset.description, preset.base_url]
+    .some(value => value?.toLocaleLowerCase().includes(query)))
+})
+</script>
+
+<template>
+  <div class="preset-selector">
+    <label class="field" for="provider-search"><span>提供商预设</span><input id="provider-search" v-model="search" class="input" type="search" placeholder="搜索提供商，例如 通义千问 / DeepSeek" /></label>
+    <div class="preset-grid" role="group" aria-label="提供商预设">
+      <button type="button" class="preset-chip" :class="{ selected: !modelValue }" :aria-pressed="!modelValue" @click="emit('update:modelValue', '')"><ProviderLogo /><span>自定义</span></button>
+      <button v-for="preset in filtered" :key="preset.preset_id" type="button" class="preset-chip" :class="{ selected: modelValue === preset.preset_id }" :aria-pressed="modelValue === preset.preset_id" :title="preset.description || preset.name" :data-preset="preset.preset_id" @click="emit('update:modelValue', preset.preset_id)">
+        <ProviderLogo :logo-id="preset.logo_id || preset.preset_id" /><span>{{ preset.name }}</span>
+      </button>
+    </div>
+    <p v-if="search && !filtered.length" class="subtle" role="status">没有匹配的预设，可以使用自定义服务。</p>
+  </div>
+</template>
+
+<style scoped>
+.preset-selector { display: grid; gap: var(--space-sm); }
+.preset-grid { display: flex; flex-wrap: wrap; gap: 8px; max-height: 220px; overflow-y: auto; padding: 3px; }
+.preset-chip { display: inline-flex; align-items: center; gap: 7px; padding: 6px 10px; border: 1px solid var(--color-border-default); border-radius: 11px; background: var(--color-surface-primary); color: var(--color-text-primary); cursor: pointer; font: inherit; font-size: 13px; }
+.preset-chip:hover { background: var(--color-background-hover); }
+.preset-chip.selected { border-color: #377cf6; background: color-mix(in srgb, #377cf6 12%, var(--color-surface-primary)); color: #377cf6; box-shadow: 0 0 0 1px #377cf6; }
+.preset-chip:focus-visible { outline: 2px solid var(--color-border-focus); outline-offset: 2px; }
+</style>

frontend/src/features/settings/SettingsView.vue

@@ -1,6 +1,9 @@
 <script setup lang="ts">
-import { computed, onMounted, reactive, ref } from 'vue'
+import { onMounted, ref } from 'vue'
-import type { ProviderConfig, ProviderType } from '@/contracts'
+import type { ProviderConfig } from '@/contracts'
+import ProviderForm from './ProviderForm.vue'
+import ProviderLogo from './ProviderLogo.vue'
+import ModelRoutingSettings from './ModelRoutingSettings.vue'
 import { useProviderStore } from '@/stores/provider'
 import { useSettingsStore } from '@/stores/settings'
 import { useThemeStore } from '@/stores/theme'
@@ -15,14 +18,9 @@ const settingsStore = useSettingsStore()
 const providerStore = useProviderStore()
 const themeStore = useThemeStore()
 const showProviderForm = ref(false)
-const editingProviderId = ref<string | null>(null)
+const editingProvider = ref<ProviderConfig>()
 const providerAction = ref('')
 const testResults = ref<Record<string, string>>({})
-const providerApiKey = ref('')
-const providerForm = reactive({ preset_id: '', provider_type: 'openai_compatible' as ProviderType, name: '', base_url: '', default_model: '', credential_id: '', enabled: true })
-const formModels = computed(() => editingProviderId.value ? providerStore.modelsByProvider[editingProviderId.value] ?? [] : [])
-const selectedPreset = computed(() => providerStore.presets.find((item) => item.preset_id === providerForm.preset_id) ?? null)
-
 onMounted(async () => {
   await Promise.all([providerStore.loadProviders(), providerStore.loadPresets(), settingsStore.loadDiagnostics()])
   await providerStore.refreshEnabledModels()
@@ -36,54 +34,15 @@ function presetIdFor(provider?: ProviderConfig) {
 }
 
 function openProvider(provider?: ProviderConfig) {
-  editingProviderId.value = provider?.provider_id ?? null
+  editingProvider.value = provider
-  const presetId = presetIdFor(provider)
+  providerAction.value = ''
-  const preset = providerStore.presets.find((item) => item.preset_id === presetId)
-  Object.assign(providerForm, { preset_id: presetId, provider_type: provider?.provider_type ?? 'openai_compatible', name: provider?.name ?? '', base_url: provider?.base_url ?? '', default_model: provider?.default_model ?? '', credential_id: provider?.credential_id ?? preset?.default_credential_id ?? '', enabled: provider?.enabled ?? true })
-  providerApiKey.value = ''
   showProviderForm.value = true
-  if (providerForm.credential_id) void providerStore.loadCredentialStatus(providerForm.credential_id).catch(() => undefined)
   if (provider) void providerStore.loadModels(provider.provider_id).catch(() => undefined)
 }
 
-function applyProviderPreset() {
+async function providerSaved(provider: ProviderConfig) {
-  const preset = providerStore.presets.find((item) => item.preset_id === providerForm.preset_id)
+  await providerStore.loadProviders()
-  if (!preset) return
+  if (provider.enabled) void providerStore.loadModels(provider.provider_id).catch(() => undefined)
-  Object.assign(providerForm, {
-    provider_type: preset.provider_type,
-    name: preset.name,
-    base_url: preset.base_url,
-    credential_id: preset.default_credential_id ?? '',
-  })
-  providerApiKey.value = ''
-  if (providerForm.credential_id) void providerStore.loadCredentialStatus(providerForm.credential_id).catch(() => undefined)
-}
-
-function closeProvider() {
-  providerApiKey.value = ''
-  showProviderForm.value = false
-}
-
-async function saveProvider() {
-  providerAction.value = ''
-  const credentialId = providerForm.credential_id.trim()
-  const requiresApiKey = Boolean(selectedPreset.value?.requires_credential)
-  if (requiresApiKey && !providerApiKey.value && !providerStore.credentialConfiguredById[credentialId]) {
-    providerAction.value = '请输入 API Key。密钥将由后端加密保存。'
-    return
-  }
-  const data = { ...providerForm, base_url: providerForm.base_url || undefined, credential_id: providerForm.credential_id || undefined, capabilities: {}, has_credential: Boolean(providerForm.credential_id) }
-  try {
-    if (providerApiKey.value) await providerStore.saveCredential(credentialId, providerApiKey.value)
-    const saved = editingProviderId.value
-      ? await providerStore.updateProvider(editingProviderId.value, data)
-      : await providerStore.addProvider(data)
-    closeProvider()
-    if (saved.enabled) void providerStore.loadModels(saved.provider_id).catch(() => undefined)
-  } catch (error) {
-    providerApiKey.value = ''
-    providerAction.value = error instanceof Error ? error.message : 'Provider 保存失败'
-  }
 }
 
 async function removeProvider(provider: ProviderConfig) { if (!confirm(`确定删除 Provider“${provider.name}”吗？`)) return; try { await providerStore.deleteProvider(provider.provider_id) } catch (error) { providerAction.value = error instanceof Error ? error.message : '删除失败' } }
@@ -107,14 +66,15 @@ async function chooseDefaultModel(provider: ProviderConfig, event: Event) {
 
     <div v-else-if="activeSection === 'providers'" class="settings-section">
       <div class="section-head">
-        <div><h2>模型提供商</h2><p class="subtle">支持 OpenAI、DeepSeek、Ollama 和自定义兼容服务。</p></div>
+        <div><h2>模型提供商</h2><p class="subtle">选择国内外提供商预设，或配置自定义 API 与独立密钥。</p></div>
         <button class="button-primary" @click="openProvider()">新增 Provider</button>
       </div>
       <div v-if="providerStore.error || providerAction" class="error-banner">{{ providerStore.error || providerAction }}</div>
+      <p v-if="!providerStore.providers.length" class="subtle">{{ providerStore.isLoading ? '正在加载提供商…' : '尚无可用提供商，请添加真实 API 或本地 Ollama 配置。' }}</p>
       <div class="provider-list">
         <article v-for="provider in providerStore.providers" :key="provider.provider_id" class="item-card provider-card">
           <div class="provider-main">
-            <div class="inline-actions"><strong>{{ provider.name }}</strong><span class="badge" :class="{ success: provider.enabled }">{{ provider.provider_type }}</span></div>
+            <div class="inline-actions"><ProviderLogo :logo-id="providerStore.presets.find(preset => preset.preset_id === presetIdFor(provider))?.logo_id || presetIdFor(provider)" /><strong>{{ provider.name }}</strong><span class="badge" :class="{ success: provider.enabled }">{{ provider.provider_type }}</span></div>
             <p class="subtle">{{ provider.base_url || '本地内置' }} · 默认模型 {{ provider.default_model || '未设置' }}</p>
             <div class="tag-list"><span v-for="(_, capability) in provider.capabilities" :key="capability" class="badge">{{ capability }}</span></div>
             <div v-if="providerStore.modelsByProvider[provider.provider_id]?.length" class="model-picker">
@@ -132,48 +92,19 @@ async function chooseDefaultModel(provider: ProviderConfig, event: Event) {
             <button class="button-secondary" :disabled="providerStore.modelLoadingByProvider[provider.provider_id]" @click="refreshModels(provider)">{{ providerStore.modelLoadingByProvider[provider.provider_id] ? '获取中…' : '刷新模型' }}</button>
             <button class="button-secondary" @click="testProvider(provider)">测试</button>
             <button class="button-secondary" @click="openProvider(provider)">编辑</button>
-            <button class="button-danger" :disabled="provider.provider_id === 'mock'" @click="removeProvider(provider)">删除</button>
+            <button class="button-danger" @click="removeProvider(provider)">删除</button>
           </div>
         </article>
       </div>
     </div>
 
-    <div v-else-if="activeSection === 'index'" class="panel settings-section"><h2>索引与模型</h2><div class="index-summary"><div><span class="badge" :class="{ success: settingsStore.indexStatus.status === 'idle', error: settingsStore.indexStatus.status === 'error' }">{{ settingsStore.indexStatus.status }}</span><p>待处理任务 {{ settingsStore.indexStatus.pending_jobs }}</p></div><div><strong>{{ settingsStore.indexStatus.total_notes }}</strong><small>笔记</small></div><div><strong>{{ settingsStore.indexStatus.total_blocks }}</strong><small>Block</small></div></div><div v-if="settingsStore.indexStatus.error" class="error-banner">{{ settingsStore.indexStatus.error }}</div><div class="inline-actions"><button class="button-primary" @click="settingsStore.rebuildIndex('full')">重建全部</button><button class="button-secondary" @click="settingsStore.rebuildIndex('fts')">重建文本索引</button><button class="button-secondary" @click="settingsStore.rebuildIndex('vector')">重建向量索引</button></div></div>
+    <div v-else-if="activeSection === 'index'" class="panel settings-section"><h2>索引与模型</h2><div class="index-summary"><div><span class="badge" :class="{ success: settingsStore.indexStatus.status === 'idle', error: settingsStore.indexStatus.status === 'error' }">{{ settingsStore.indexStatus.status }}</span><p>待处理任务 {{ settingsStore.indexStatus.pending_jobs }}</p></div><div><strong>{{ settingsStore.indexStatus.total_notes ?? '未获取' }}</strong><small>笔记</small></div><div><strong>{{ settingsStore.indexStatus.total_blocks ?? '未获取' }}</strong><small>Block</small></div></div><div v-if="settingsStore.indexStatus.error" class="error-banner">{{ settingsStore.indexStatus.error }}</div><div class="inline-actions"><button class="button-primary" @click="settingsStore.rebuildIndex('full')">重建全部</button><span class="subtle">当前后端支持全量重建。</span></div><ModelRoutingSettings /></div>
 
-    <div v-else-if="activeSection === 'permissions'" class="panel settings-section"><h2>权限策略</h2><p class="muted section-description">高影响能力默认需要确认。未知权限由后端拒绝。</p><div class="permission-list"><div v-for="(policy, permission) in settingsStore.permissionPolicy" :key="permission" class="setting-row"><span><strong>{{ permission }}</strong></span><select :value="policy" class="select short" @change="settingsStore.setPermission(String(permission), ($event.target as HTMLSelectElement).value as 'allow' | 'confirm' | 'deny')"><option value="allow">允许</option><option value="confirm">每次确认</option><option value="deny">拒绝</option></select></div></div></div>
+    <div v-else-if="activeSection === 'permissions'" class="panel settings-section"><h2>权限策略</h2><p class="muted section-description">以下为后端当前生效的权限策略；全局策略编辑尚未开放，运行时按实际权限请求确认。</p><p v-if="!Object.keys(settingsStore.permissionPolicy).length" class="subtle">尚未获取权限策略，请检查后端连接并重新检测。</p><div class="permission-list"><div v-for="(policy, permission) in settingsStore.permissionPolicy" :key="permission" class="setting-row"><span><strong>{{ permission }}</strong></span><span>{{ policy === 'allow' ? '允许' : policy === 'confirm' ? '每次确认' : '拒绝' }}</span></div></div></div>
 
-    <div v-else class="panel settings-section"><h2>AI Core 诊断</h2><div v-if="settingsStore.diagnosticsError" class="error-banner">{{ settingsStore.diagnosticsError }}</div><div class="diagnostic-grid"><div class="item-card"><span class="badge" :class="{ success: settingsStore.aiCoreStatus === 'running', error: settingsStore.aiCoreStatus === 'error' }">{{ settingsStore.aiCoreStatus }}</span><h3>Sidecar 状态</h3><p class="subtle">AI Core 不可用时，Markdown 编辑仍可继续使用。</p></div><div class="item-card"><strong>{{ settingsStore.aiCoreAddress }}</strong><h3>开发 API 地址</h3><p class="subtle">正式桌面环境由 Sidecar Manager 动态提供。</p></div></div><div class="inline-actions diagnostic-actions"><button class="button-primary" @click="settingsStore.loadDiagnostics">重新检测</button><button class="button-secondary" @click="settingsStore.restartAiCore">重启 AI Core</button></div></div>
+    <div v-else class="panel settings-section"><h2>AI Core 诊断</h2><div v-if="settingsStore.diagnosticsError" class="error-banner">{{ settingsStore.diagnosticsError }}</div><div class="diagnostic-grid"><div class="item-card"><span class="badge" :class="{ success: settingsStore.aiCoreStatus === 'running', error: settingsStore.aiCoreStatus === 'error' }">{{ settingsStore.aiCoreStatus }}</span><h3>AI Core 连接状态</h3><p class="subtle">AI Core 不可用时，Markdown 编辑仍可继续使用。</p></div><div class="item-card"><strong>{{ settingsStore.aiCoreAddress }}</strong><h3>开发 API 地址</h3><p class="subtle">正式桌面环境由 Sidecar Manager 动态提供。</p></div></div><div class="inline-actions diagnostic-actions"><button class="button-primary" @click="settingsStore.loadDiagnostics">重新检测</button><span class="subtle">当前 Web 端不支持重启后端进程，请在运行后端的终端中操作。</span></div></div>
 
-    <div v-if="showProviderForm" class="modal-backdrop" @click.self="closeProvider">
+    <ProviderForm v-if="showProviderForm" :provider="editingProvider" :models="editingProvider ? providerStore.modelsByProvider[editingProvider.provider_id] : []" @close="showProviderForm = false" @saved="providerSaved" />
-      <div class="modal">
-        <h2>{{ editingProviderId ? '编辑 Provider' : '新增 Provider' }}</h2>
-        <form @submit.prevent="saveProvider">
-          <div class="field">
-            <label>提供商预设</label>
-            <select v-model="providerForm.preset_id" class="select" @change="applyProviderPreset">
-              <option value="">自定义</option>
-              <option v-for="preset in providerStore.presets" :key="preset.preset_id" :value="preset.preset_id">{{ preset.name }}</option>
-            </select>
-          </div>
-          <div class="field"><label>接入协议</label><select v-model="providerForm.provider_type" class="select"><option value="openai_compatible">OpenAI Compatible</option><option value="openai_chat">OpenAI Chat</option><option value="openai_responses">OpenAI Responses</option><option value="anthropic_messages">Anthropic Messages</option><option value="ollama">Ollama</option></select></div>
-          <div class="field"><label>名称</label><input v-model="providerForm.name" class="input" required /></div>
-          <div class="field"><label>Base URL</label><input v-model="providerForm.base_url" class="input" placeholder="https://api.example.com/v1" required /></div>
-          <div class="field">
-            <label>默认模型</label>
-            <input v-model="providerForm.default_model" class="input" :list="editingProviderId ? 'provider-model-options' : undefined" placeholder="保存后自动获取，也可以手动输入" />
-            <datalist id="provider-model-options"><option v-for="model in formModels" :key="model.model_id" :value="model.model_id">{{ model.name }}</option></datalist>
-          </div>
-          <div v-if="selectedPreset?.requires_credential" class="field">
-            <label>API Key</label>
-            <input v-model="providerApiKey" class="input" type="password" autocomplete="new-password" spellcheck="false" :placeholder="providerStore.credentialConfiguredById[providerForm.credential_id] ? '已配置，留空表示不修改' : '请输入 API Key'" />
-            <small class="subtle">提交后由本地 AI Core 加密保存，页面不会回显已保存的密钥。</small>
-          </div>
-          <div v-else-if="!selectedPreset" class="field"><label>Credential ID</label><input v-model="providerForm.credential_id" class="input" placeholder="自定义凭据标识" /><small class="subtle">自定义服务可以引用 Host 注入或后端已保存的凭据。</small></div>
-          <label class="inline-actions"><input v-model="providerForm.enabled" type="checkbox" /> 启用</label>
-          <div class="inline-actions"><button class="button-primary">保存并获取模型</button><button type="button" class="button-secondary" @click="closeProvider">取消</button></div>
-        </form>
-      </div>
-    </div>
   </section>
 </template>
 

frontend/src/features/skills/SkillsView.vue

@@ -32,6 +32,7 @@ async function uninstall(skillId: string, name: string) {
       <div class="detail-grid"><div><h3>工具</h3><div class="tag-list"><span v-for="tool in skillStore.selectedSkill.tools" :key="tool" class="badge info">{{ tool }}</span></div></div><div><h3>权限</h3><div class="tag-list"><span v-for="permission in skillStore.selectedSkill.permissions" :key="permission" class="badge warning">{{ permission }}</span></div></div><div><h3>检索配置</h3><pre>{{ JSON.stringify(skillStore.selectedSkill.retrieval_config, null, 2) }}</pre></div><div><h3>模型能力</h3><div class="tag-list"><span v-for="cap in skillStore.selectedSkill.model_requirements?.capabilities" :key="cap" class="badge">{{ cap }}</span></div></div></div>
       <div v-if="skillStore.selectedSkill.missing_dependencies?.length" class="error-banner dependencies">缺失依赖：{{ skillStore.selectedSkill.missing_dependencies.join('、') }}</div>
     </div>
+    <div v-else-if="!skillStore.skills.length" class="empty-state"><div><strong>{{ skillStore.isLoading ? '正在加载…' : skillStore.error ? '加载失败' : '尚未安装' }}</strong><button class="button-secondary" @click="skillStore.loadSkills">重新加载</button></div></div>
     <div v-else class="feature-grid"><article v-for="skill in skillStore.skills" :key="skill.skill_id" class="item-card extension-card" @click="skillStore.selectSkill(skill.skill_id)"><div class="extension-title"><AppIcon :icon="Lightning" :size="22" /><div><strong>{{ skill.name }}</strong><p>v{{ skill.version }}</p></div><span class="badge" :class="{ success: skill.status === 'ready', warning: skill.status === 'dependency_missing' }">{{ skill.status }}</span></div><p class="muted">{{ skill.description }}</p><div class="tag-list"><span v-for="permission in skill.permissions.slice(0, 3)" :key="permission" class="badge">{{ permission }}</span></div></article></div>
   </section>
 </template>

frontend/src/features/vault/VaultEntry.vue

@@ -23,13 +23,13 @@ onMounted(async () => {
       await openVault(lastVaultPath)
       return
     } catch {
-      // Mock 阶段保存的旧路径可能与当前后端 Vault 不同，清除后让用户重新选择。
+      // 历史保存的旧路径可能与当前后端 Vault 不同，清除后让用户重新选择。
       localStorage.removeItem('last-vault-path')
     }
   }
-  setTimeout(() => {
+  {
     aiCoreStatus.value = settingsStore.aiCoreStatus === 'running' ? 'running' : 'stopped'
-  }, 800)
+  }
 })
 
 async function openVault(path: string) {

frontend/src/services/agentService.ts

@@ -1,6 +1,6 @@
 import apiClient from './apiClient'
 import { SseClient } from './sseClient'
-import type { AgentRun, AgentEvent, AgentTraceResponse, ApiAgentRun, OperationResponse, PageMeta, ToolDefinition, PermissionRequest } from '@/contracts'
+import type { AgentRun, AgentEvent, AgentTraceResponse, ApiAgentRun, OperationResponse, PageMeta, ToolDefinition } from '@/contracts'
 
 function toAgentRun(run: ApiAgentRun): AgentRun {
   // API 的 token_usage 是累计值，UI 模型预留了输入/输出拆分字段。
@@ -10,8 +10,6 @@ function toAgentRun(run: ApiAgentRun): AgentRun {
     current_step: run.current_step,
     max_steps: run.max_steps,
     token_usage: {
-      input_tokens: 0,
-      output_tokens: 0,
       total_tokens: run.token_usage,
     },
     started_at: run.created_at,
@@ -107,207 +105,3 @@ export async function respondToPermission(
     decision,
   })
 }
-
-export const mockTools: ToolDefinition[] = [
-  {
-    name: 'notes.search',
-    description: '搜索笔记，支持关键词和语义检索',
-    parameters: {
-      type: 'object',
-      properties: {
-        query: { type: 'string', description: '搜索关键词' },
-        limit: { type: 'number', description: '返回结果数量' },
-      },
-      required: ['query'],
-    },
-    source: 'builtin',
-  },
-  {
-    name: 'notes.read',
-    description: '读取指定笔记的完整内容',
-    parameters: {
-      type: 'object',
-      properties: {
-        note_id: { type: 'string' },
-      },
-      required: ['note_id'],
-    },
-    source: 'builtin',
-  },
-  {
-    name: 'notes.create',
-    description: '创建新笔记',
-    parameters: {
-      type: 'object',
-      properties: {
-        title: { type: 'string' },
-        content: { type: 'string' },
-        folder_path: { type: 'string' },
-      },
-      required: ['title', 'content'],
-    },
-    source: 'builtin',
-  },
-  {
-    name: 'rag.search',
-    description: '基于 RAG 的语义检索，返回相关知识片段',
-    parameters: {
-      type: 'object',
-      properties: {
-        query: { type: 'string' },
-        top_k: { type: 'number' },
-      },
-      required: ['query'],
-    },
-    source: 'builtin',
-  },
-  {
-    name: 'tasks.create',
-    description: '创建任务',
-    parameters: {
-      type: 'object',
-      properties: {
-        title: { type: 'string' },
-        description: { type: 'string' },
-        priority: { type: 'string', enum: ['low', 'medium', 'high'] },
-      },
-      required: ['title'],
-    },
-    source: 'builtin',
-  },
-  {
-    name: 'system.echo',
-    description: '回显输入内容（测试用）',
-    parameters: {
-      type: 'object',
-      properties: {
-        text: { type: 'string' },
-      },
-      required: ['text'],
-    },
-    source: 'builtin',
-  },
-  {
-    name: 'math.add',
-    description: '两数相加（测试用）',
-    parameters: {
-      type: 'object',
-      properties: {
-        a: { type: 'number' },
-        b: { type: 'number' },
-      },
-      required: ['a', 'b'],
-    },
-    source: 'builtin',
-  },
-]
-
-export const mockAgentRuns: AgentRun[] = [
-  {
-    run_id: 'run-1',
-    status: 'completed',
-    current_step: 3,
-    max_steps: 10,
-    token_usage: { input_tokens: 2340, output_tokens: 890, total_tokens: 3230 },
-    started_at: '2026-08-25T11:00:00Z',
-    completed_at: '2026-08-25T11:02:30Z',
-  },
-  {
-    run_id: 'run-2',
-    status: 'running',
-    current_step: 2,
-    max_steps: 10,
-    token_usage: { input_tokens: 1500, output_tokens: 420, total_tokens: 1920 },
-    started_at: '2026-08-26T09:30:00Z',
-  },
-]
-
-export const mockAgentEvents: AgentEvent[] = [
-  {
-    event: 'RunStarted',
-    sequence: 1,
-    run_id: 'run-1',
-    data: { task: '帮我整理红黑树的核心知识点' },
-    timestamp: '2026-08-25T11:00:00Z',
-  },
-  {
-    event: 'ThinkingDelta',
-    sequence: 2,
-    run_id: 'run-1',
-    data: { text: '我需要先搜索笔记中关于红黑树的内容...' },
-    timestamp: '2026-08-25T11:00:01Z',
-  },
-  {
-    event: 'ToolCall',
-    sequence: 3,
-    run_id: 'run-1',
-    data: {
-      tool_call_id: 'tc-1',
-      name: 'notes.search',
-      parameters: { query: '红黑树 插入 删除', limit: 5 },
-      status: 'running',
-    },
-    timestamp: '2026-08-25T11:00:02Z',
-  },
-  {
-    event: 'ToolResult',
-    sequence: 4,
-    run_id: 'run-1',
-    data: {
-      tool_call_id: 'tc-1',
-      name: 'notes.search',
-      status: 'completed',
-      result: '找到 5 条相关结果，包括红黑树性质、插入操作、删除操作等...',
-      duration_ms: 320,
-    },
-    timestamp: '2026-08-25T11:00:02Z',
-  },
-  {
-    event: 'Citation',
-    sequence: 5,
-    run_id: 'run-1',
-    data: {
-      note_id: 'n-rbt',
-      block_id: 'b1',
-      heading_path: '数据结构 / 红黑树 / 性质',
-    },
-    timestamp: '2026-08-25T11:00:03Z',
-  },
-  {
-    event: 'ThinkingDelta',
-    sequence: 6,
-    run_id: 'run-1',
-    data: { text: '搜索结果很全面，让我整理一下结构...' },
-    timestamp: '2026-08-25T11:00:03Z',
-  },
-  {
-    event: 'TextDelta',
-    sequence: 7,
-    run_id: 'run-1',
-    data: { text: '## 红黑树核心知识点整理\n\n### 1. 基本性质\n红黑树是一种自平衡二叉搜索树，每个节点带有颜色属性...' },
-    timestamp: '2026-08-25T11:00:04Z',
-  },
-  {
-    event: 'Usage',
-    sequence: 8,
-    run_id: 'run-1',
-    data: { input_tokens: 2340, output_tokens: 890, total_tokens: 3230 },
-    timestamp: '2026-08-25T11:02:30Z',
-  },
-  {
-    event: 'RunCompleted',
-    sequence: 9,
-    run_id: 'run-1',
-    data: { message: 'Task completed successfully' },
-    timestamp: '2026-08-25T11:02:30Z',
-  },
-]
-
-export const mockPermissionRequest: PermissionRequest = {
-  request_id: 'perm-1',
-  run_id: 'run-2',
-  tool_name: 'notes.create',
-  permission: 'notes.write',
-  parameters: { title: '红黑树知识点总结', folder_path: '/数据结构' },
-  impact: '将在你的知识库中创建一篇新笔记',
-}

frontend/src/services/chatService.ts

@@ -1,5 +1,5 @@
 import { SseClient } from './sseClient'
-import type { Conversation, ChatMessage, ModelEvent } from '@/contracts'
+import type { ModelEvent } from '@/contracts'
 
 export interface ChatRequest {
   provider_id: string
@@ -46,87 +46,3 @@ export function streamChat(
   client.connect().catch(() => {})
   return client
 }
-
-export const mockConversations: Conversation[] = [
-  {
-    conversation_id: 'conv-1',
-    title: '关于红黑树的讨论',
-    created_at: '2026-08-25T10:00:00Z',
-    updated_at: '2026-08-25T10:30:00Z',
-    message_count: 6,
-  },
-  {
-    conversation_id: 'conv-2',
-    title: '死锁避免算法',
-    created_at: '2026-08-24T14:00:00Z',
-    updated_at: '2026-08-24T15:20:00Z',
-    message_count: 4,
-  },
-  {
-    conversation_id: 'conv-3',
-    title: 'TCP三次握手',
-    created_at: '2026-08-22T09:00:00Z',
-    updated_at: '2026-08-22T09:15:00Z',
-    message_count: 3,
-  },
-]
-
-export const mockMessages: Record<string, ChatMessage[]> = {
-  'conv-1': [
-    {
-      message_id: 'msg-1',
-      conversation_id: 'conv-1',
-      role: 'user',
-      content: '红黑树的插入操作有哪些情况需要调整？',
-      created_at: '2026-08-25T10:00:00Z',
-    },
-    {
-      message_id: 'msg-2',
-      conversation_id: 'conv-1',
-      role: 'assistant',
-      content:
-        '红黑树插入后可能违反红黑性质，需要通过旋转和变色来修复。主要有以下几种情况：\n\n**情况1：叔叔节点是红色**\n- 将父节点和叔叔节点设为黑色\n- 将祖父节点设为红色\n- 当前节点上移到祖父节点，继续向上调整\n\n**情况2：叔叔节点是黑色，且当前节点是右孩子**\n- 以父节点为支点左旋\n- 将当前节点转换为左孩子，进入情况3\n\n**情况3：叔叔节点是黑色，且当前节点是左孩子**\n- 以祖父节点为支点右旋\n- 将父节点设为黑色，祖父节点设为红色\n\n以上三种情况需要从插入节点开始向上递归处理，直到根节点。',
-      created_at: '2026-08-25T10:01:00Z',
-      citations: [
-        {
-          note_id: 'n-rbt',
-          block_id: 'b1',
-          file_path: '/数据结构/红黑树.md',
-          heading_path: '数据结构 / 红黑树 / 插入操作',
-          content: '插入后可能破坏红黑性质，需要通过变色和旋转来修复...',
-        },
-        {
-          note_id: 'n-rbt',
-          block_id: 'b2',
-          file_path: '/数据结构/红黑树.md',
-          heading_path: '数据结构 / 红黑树 / 五种情况',
-          content: 'Case 1: 叔叔为红色，直接变色...',
-        },
-      ],
-    },
-    {
-      message_id: 'msg-3',
-      conversation_id: 'conv-1',
-      role: 'user',
-      content: '能对比一下 AVL 树和红黑树的性能差异吗？',
-      created_at: '2026-08-25T10:10:00Z',
-    },
-    {
-      message_id: 'msg-4',
-      conversation_id: 'conv-1',
-      role: 'assistant',
-      content:
-        'AVL 树和红黑树都是平衡二叉搜索树，但它们的平衡策略和性能特点有所不同：\n\n| 特性 | AVL 树 | 红黑树 |\n|------|--------|--------|\n| 严格平衡 | 高度差 ≤ 1 | 黑色高度相同 |\n| 查找速度 | 更快（严格平衡） | 略慢但仍为 O(log n) |\n| 插入/删除 | 旋转更多 | 旋转更少（颜色调整为主） |\n| 存储开销 | 每个节点存高度（int） | 每个节点存颜色（1 bit） |\n| 适用场景 | 读多写少 | 读写均衡或写较多 |\n\n**总结：**\n- 如果应用以查询为主，很少修改，AVL 树可能更优\n- 如果插入删除频繁，红黑树更合适，重平衡开销更低\n- 红黑树在工业界应用更广泛（C++ STL 的 map/set、Java 的 TreeMap 等）',
-      created_at: '2026-08-25T10:11:00Z',
-      citations: [
-        {
-          note_id: 'n-rbt',
-          block_id: 'b3',
-          file_path: '/数据结构/红黑树.md',
-          heading_path: '数据结构 / 红黑树 / 与AVL树对比',
-          content: '红黑树相比AVL树，牺牲了部分平衡性以换取更少的旋转操作...',
-        },
-      ],
-    },
-  ],
-}

frontend/src/services/index.ts

@@ -10,6 +10,7 @@ export * as skillService from './skillService'
 export * as pluginService from './pluginService'
 export * as mcpServerService from './mcpServerService'
 export * as providerService from './providerService'
+export * as modelRoutingService from './modelRoutingService'
 export * as taskService from './taskService'
 export * as indexService from './indexService'
 export * as systemService from './systemService'

frontend/src/services/indexService.ts

@@ -5,10 +5,8 @@ function toIndexStatus(status: ApiIndexStatus): IndexStatus {
   return {
     status: status.status === 'idle' ? 'idle' : status.status === 'failed' ? 'error' : 'indexing',
     pending_jobs: status.pending_jobs,
-    total_notes: 0,
+    total_notes: status.total_notes ?? null,
-    total_blocks: 0,
+    total_blocks: status.total_blocks ?? null,
-    fts_enabled: true,
-    vector_enabled: true,
     last_indexed_at: status.last_completed_at ?? undefined,
     error: status.error_message ?? undefined,
   }
@@ -26,15 +24,3 @@ export async function rebuildIndex(scope: 'full' | 'fts' | 'vector' = 'full'): P
 export async function getIndexJob(jobId: string): Promise<ApiIndexJob> {
   return apiClient.get(`/api/index/jobs/${jobId}`)
 }
-
-export const mockIndexStatus: IndexStatus = {
-  status: 'idle',
-  pending_jobs: 0,
-  total_notes: 42,
-  total_blocks: 318,
-  fts_enabled: true,
-  vector_enabled: true,
-  embedding_model: 'bge-m3',
-  reranker_model: 'bge-reranker-base',
-  last_indexed_at: new Date().toISOString(),
-}

frontend/src/services/modelRoutingService.spec.ts

@@ -0,0 +1,28 @@
+// @vitest-environment happy-dom
+import { afterEach, beforeEach, describe, expect, it, vi } from 'vitest'
+import { getModelRouting, saveModelRouting } from './modelRoutingService'
+
+const config = { version: 7, embedding: { provider_id: 'p1', model: 'embedding', endpoint: '/embeddings', dimensions: 1024 }, transcription: null, speaker_matching: null }
+const result = { config, local_backends: [{ capability: 'embedding', status: 'placeholder', message: 'hash' }] }
+const json = (body: unknown, status = 200) => new Response(JSON.stringify(body), { status, headers: { 'Content-Type': 'application/json' } })
+beforeEach(() => { vi.stubGlobal('fetch', vi.fn()) })
+afterEach(() => { vi.unstubAllGlobals() })
+
+describe('model routing service', () => {
+  it('round-trips versioned routing without an extra config wrapper', async () => {
+    vi.mocked(fetch).mockImplementation(async () => json(result))
+    expect(await getModelRouting()).toEqual(result)
+    expect(await saveModelRouting(config)).toEqual(result)
+    expect(fetch).toHaveBeenNthCalledWith(1, '/api/model-routing', expect.objectContaining({ method: 'GET' }))
+    expect(fetch).toHaveBeenNthCalledWith(2, '/api/model-routing', expect.objectContaining({ method: 'PUT', body: JSON.stringify(config) }))
+  })
+
+  it('surfaces load, save and version conflict errors instead of returning local defaults', async () => {
+    vi.mocked(fetch).mockRejectedValueOnce(new Error('offline'))
+      .mockResolvedValueOnce(json({ error: { code: 'MODEL_ROUTING_VERSION_CONFLICT', message: 'conflict' } }, 409))
+      .mockResolvedValueOnce(json({ error: { code: 'SAVE_FAILED', message: 'disk full' } }, 500))
+    await expect(getModelRouting()).rejects.toMatchObject({ code: 'NETWORK_ERROR' })
+    await expect(saveModelRouting(config)).rejects.toMatchObject({ code: 'MODEL_ROUTING_VERSION_CONFLICT' })
+    await expect(saveModelRouting(config)).rejects.toMatchObject({ code: 'SAVE_FAILED' })
+  })
+})

frontend/src/services/modelRoutingService.ts

@@ -0,0 +1,11 @@
+import type { ModelRoutingConfig, ModelRoutingResponse } from '@/contracts'
+import apiClient from './apiClient'
+
+export function getModelRouting(): Promise<ModelRoutingResponse> {
+  return apiClient.get('/api/model-routing')
+}
+
+// version is the last version read from the server (optimistic concurrency).
+export function saveModelRouting(config: ModelRoutingConfig): Promise<ModelRoutingResponse> {
+  return apiClient.put('/api/model-routing', config)
+}

frontend/src/services/pluginService.ts

@@ -126,92 +126,3 @@ export async function deletePluginSecret(pluginId: string, key: string): Promise
 export async function uninstallPlugin(pluginId: string): Promise<OperationResponse> {
   return apiClient.delete(`/api/plugins/${pluginId}`)
 }
-
-export const mockPlugins: Plugin[] = [
-  {
-    plugin_id: 'github-integration',
-    name: 'GitHub 集成',
-    version: '1.3.2',
-    description: '接入 GitHub API，支持搜索 Issue、查看 PR 和管理仓库',
-    icon: '',
-    author: 'NotesAgent 团队',
-    status: 'ready',
-    enabled: true,
-    permissions: ['notes.read', 'network.request'],
-    contributions: [
-      { type: 'tool', id: 'github.search_issues', name: '搜索 Issue', description: '搜索 GitHub 仓库中的 Issue' },
-      { type: 'tool', id: 'github.get_pr', name: '获取 PR 详情', description: '获取 Pull Request 的详细信息' },
-      { type: 'command', id: 'github.open_repo', name: '打开仓库', description: '在浏览器中打开对应 GitHub 仓库' },
-    ],
-    backend_type: 'mcp',
-    transport: 'stdio',
-    dependent_skills: ['research-assistant'],
-  },
-  {
-    plugin_id: 'translator',
-    name: '翻译助手',
-    version: '1.0.0',
-    description: '提供多语言翻译能力，支持文档批量翻译',
-    icon: '',
-    author: '社区贡献',
-    status: 'ready',
-    enabled: false,
-    permissions: ['notes.read', 'notes.write', 'network.request'],
-    contributions: [
-      { type: 'tool', id: 'translator.translate', name: '翻译文本', description: '翻译指定文本到目标语言' },
-      { type: 'command', id: 'translator.translate_note', name: '翻译当前笔记', description: '翻译当前打开的笔记' },
-      { type: 'settings_section', id: 'translator.settings', name: '翻译设置', description: '配置翻译服务和默认语言' },
-    ],
-    backend_type: 'mcp',
-    transport: 'stdio',
-  },
-  {
-    plugin_id: 'kanban',
-    name: '看板视图',
-    version: '0.8.0',
-    description: '为任务提供看板视图，支持拖拽排序和多维度筛选',
-    icon: '',
-    author: '社区贡献',
-    status: 'installed',
-    enabled: false,
-    permissions: ['tasks.read', 'tasks.write'],
-    contributions: [
-      { type: 'sidebar_panel', id: 'kanban.panel', name: '任务看板', description: '以看板方式查看和管理任务' },
-    ],
-    backend_type: 'internal_rpc',
-  },
-  {
-    plugin_id: 'pdf-importer',
-    name: 'PDF 导入',
-    version: '2.1.0',
-    description: '导入 PDF 文档，提取文本和目录结构生成笔记',
-    icon: '',
-    author: 'NotesAgent 团队',
-    status: 'error',
-    enabled: false,
-    permissions: ['notes.write', 'attachments.read'],
-    contributions: [
-      { type: 'importer', id: 'pdf.import', name: 'PDF 导入器', description: '从 PDF 文件导入内容' },
-    ],
-    backend_type: 'mcp',
-    transport: 'stdio',
-    last_error: 'PDF 解析库初始化失败，请检查 Python 依赖',
-  },
-  {
-    plugin_id: 'calendar',
-    name: '日历同步',
-    version: '0.5.0',
-    description: '同步日历事件，自动生成相关笔记和任务提醒',
-    icon: '',
-    author: '社区贡献',
-    status: 'dependency_missing',
-    enabled: false,
-    permissions: ['tasks.read', 'tasks.write', 'network.request'],
-    contributions: [
-      { type: 'tool', id: 'calendar.events', name: '日历事件', description: '获取日历事件列表' },
-      { type: 'sidebar_panel', id: 'calendar.widget', name: '日历小部件', description: '侧边栏日历视图' },
-    ],
-    backend_type: 'mcp',
-    transport: 'http',
-  },
-]

frontend/src/services/providerService.spec.ts

@@ -0,0 +1,35 @@
+// @vitest-environment happy-dom
+import { afterEach, beforeEach, describe, expect, it, vi } from 'vitest'
+import { createProvider, listProviderPresets, putCredential, updateProvider } from './providerService'
+
+const provider = { provider_id: 'provider-1', provider_type: 'openai_chat', name: 'Custom', capabilities: [], enabled: true }
+const json = (body: unknown) => new Response(JSON.stringify(body), { headers: { 'Content-Type': 'application/json' } })
+beforeEach(() => { vi.stubGlobal('fetch', vi.fn()) })
+afterEach(() => { vi.unstubAllGlobals() })
+
+describe('provider wire contracts', () => {
+  it('retains preset logos, descriptions and capabilities', async () => {
+    const preset = { preset_id: 'qwen', logo_id: 'qwen', description: '通义千问', capabilities: ['chat', 'embedding'] }
+    vi.mocked(fetch).mockResolvedValue(json({ items: [preset] }))
+    expect(await listProviderPresets()).toEqual([preset])
+  })
+
+  it('persists protocol edits and explicit credential unlinking', async () => {
+    vi.mocked(fetch).mockResolvedValue(json(provider))
+    await updateProvider('provider-1', { provider_type: 'openai_responses', default_model: '', credential_id: null })
+    expect(fetch).toHaveBeenCalledWith('/api/providers/provider-1', expect.objectContaining({ method: 'PATCH' }))
+    expect(JSON.parse(String(vi.mocked(fetch).mock.calls[0][1]?.body))).toEqual({ provider_type: 'openai_responses', default_model: '', credential_id: null })
+  })
+
+  it('sends secrets only to credentials and a reference to provider configuration', async () => {
+    vi.mocked(fetch).mockResolvedValueOnce(json({ configured: true })).mockResolvedValueOnce(json(provider))
+    await putCredential('provider-key-test', 'test-only-key')
+    await createProvider({ name: 'Custom', provider_type: 'openai_compatible', default_model: '', enabled: true, credential_id: 'provider-key-test', has_credential: true, capabilities: {} })
+    const calls = vi.mocked(fetch).mock.calls
+    expect(calls[0][0]).toBe('/api/credentials/provider-key-test')
+    expect(JSON.parse(String(calls[0][1]?.body))).toEqual({ api_key: 'test-only-key' })
+    expect(JSON.parse(String(calls[1][1]?.body))).toMatchObject({ credential_id: 'provider-key-test' })
+    expect(calls[1][1]?.body).not.toContain('test-only-key')
+    expect(calls[1][1]?.body).not.toContain('has_credential')
+  })
+})

frontend/src/services/providerService.ts

@@ -1,5 +1,5 @@
 import apiClient from './apiClient'
-import type { ApiModelInfo, ApiProviderConfig, ApiProviderPreset, ModelCapability, ModelInfo, OperationResponse, ProviderConfig, ProviderPreset } from '@/contracts'
+import type { ApiModelInfo, ApiProviderConfig, ApiProviderPreset, ModelCapability, ModelInfo, OperationResponse, ProviderConfig, ProviderPreset, ProviderUpdateRequest } from '@/contracts'
 
 function capabilityMap(capabilities: string[]): Partial<ModelCapability> {
   return Object.fromEntries(capabilities.map((capability) => [capability, true])) as Partial<ModelCapability>
@@ -15,7 +15,7 @@ function toProvider(provider: ApiProviderConfig): ProviderConfig {
     enabled: provider.enabled,
     capabilities: capabilityMap(provider.capabilities),
     credential_id: provider.credential_id ?? undefined,
-    has_credential: Boolean(provider.credential_id) || provider.provider_type === 'mock',
+    has_credential: Boolean(provider.credential_id),
   }
 }
 
@@ -25,7 +25,7 @@ function toModel(model: ApiModelInfo): ModelInfo {
 
 export async function listProviders(): Promise<ProviderConfig[]> {
   const response = await apiClient.get<{ items: ApiProviderConfig[] }>('/api/providers')
-  return response.items.map(toProvider)
+  return response.items.filter(provider => provider.provider_type !== 'mock').map(toProvider)
 }
 
 export async function getProvider(providerId: string): Promise<ProviderConfig> {
@@ -61,8 +61,9 @@ export async function putCredential(credentialId: string, apiKey: string): Promi
   )
 }
 
-export async function updateProvider(providerId: string, data: Partial<ProviderConfig>): Promise<ProviderConfig> {
+export async function updateProvider(providerId: string, data: ProviderUpdateRequest): Promise<ProviderConfig> {
   const response = await apiClient.patch<ApiProviderConfig>(`/api/providers/${providerId}`, {
+    provider_type: data.provider_type,
     name: data.name,
     base_url: data.base_url,
     default_model: data.default_model,
@@ -96,102 +97,3 @@ export async function testProvider(providerId: string): Promise<TestResult> {
     return { success: false, error_code: e.code || 'TEST_FAILED', error_message: e.message }
   }
 }
-
-export const mockProviders: ProviderConfig[] = [
-  {
-    provider_id: 'mock',
-    provider_type: 'mock',
-    name: 'Mock Provider (测试)',
-    default_model: 'mock-1',
-    enabled: true,
-    has_credential: true,
-    capabilities: {
-      chat: true,
-      tool_calling: true,
-      streaming: true,
-      vision: false,
-      reasoning: false,
-      structured_output: true,
-      embedding: false,
-    },
-  },
-  {
-    provider_id: 'openai-compat-1',
-    provider_type: 'openai_compatible',
-    name: 'OpenAI 兼容服务',
-    base_url: 'https://api.openai.com/v1',
-    default_model: 'gpt-4o-mini',
-    enabled: true,
-    has_credential: true,
-    capabilities: {
-      chat: true,
-      tool_calling: true,
-      streaming: true,
-      vision: true,
-      reasoning: false,
-      structured_output: true,
-      embedding: true,
-    },
-  },
-  {
-    provider_id: 'ollama-local',
-    provider_type: 'ollama',
-    name: 'Ollama (本地)',
-    base_url: 'http://127.0.0.1:11434',
-    default_model: 'qwen2.5:7b',
-    enabled: false,
-    has_credential: false,
-    capabilities: {
-      chat: true,
-      tool_calling: false,
-      streaming: true,
-      vision: false,
-      reasoning: false,
-      structured_output: false,
-      embedding: true,
-    },
-  },
-]
-
-export const mockModels: Record<string, ModelInfo[]> = {
-  mock: [
-    {
-      model_id: 'mock-1',
-      name: 'Mock Model v1',
-      capabilities: { chat: true, tool_calling: true, streaming: true, structured_output: true },
-      context_window: 8192,
-    },
-  ],
-  'openai-compat-1': [
-    {
-      model_id: 'gpt-4o-mini',
-      name: 'GPT-4o Mini',
-      capabilities: { chat: true, tool_calling: true, streaming: true, vision: true, structured_output: true },
-      context_window: 128000,
-    },
-    {
-      model_id: 'gpt-4o',
-      name: 'GPT-4o',
-      capabilities: { chat: true, tool_calling: true, streaming: true, vision: true, structured_output: true, reasoning: true },
-      context_window: 128000,
-    },
-    {
-      model_id: 'text-embedding-3-small',
-      name: 'Text Embedding 3 Small',
-      capabilities: { embedding: true },
-    },
-  ],
-  'ollama-local': [
-    {
-      model_id: 'qwen2.5:7b',
-      name: 'Qwen 2.5 7B',
-      capabilities: { chat: true, streaming: true },
-      context_window: 32768,
-    },
-    {
-      model_id: 'bge-m3',
-      name: 'BGE M3',
-      capabilities: { embedding: true },
-    },
-  ],
-}

frontend/src/services/searchService.ts

@@ -35,69 +35,3 @@ export async function search(request: SearchRequest): Promise<{
     mode: response.mode,
   }
 }
-
-export async function searchMock(
-  query: string,
-  mode: 'fts' | 'vector' | 'hybrid' = 'hybrid'
-): Promise<{
-  results: SearchResult[]
-  total: number
-  mode: 'fts' | 'vector' | 'hybrid'
-}> {
-  await new Promise((r) => setTimeout(r, 300))
-  if (!query.trim()) return { results: [], total: 0, mode }
-  const results: SearchResult[] = [
-    {
-      block_id: 'b1',
-      note_id: 'n-rbt',
-      note_title: '红黑树',
-      file_path: '/数据结构/红黑树.md',
-      heading_path: '数据结构 / 红黑树 / 插入操作',
-      snippet: '插入后可能破坏红黑性质，需要通过变色和旋转来修复...',
-      score: 0.95,
-      match_type: 'hybrid',
-      tags: ['数据结构', '树'],
-    },
-    {
-      block_id: 'b2',
-      note_id: 'n-rbt',
-      note_title: '红黑树',
-      file_path: '/数据结构/红黑树.md',
-      heading_path: '数据结构 / 红黑树 / 性质',
-      snippet: '红黑树是一种自平衡二叉搜索树，每个节点带有颜色属性（红或黑）...',
-      score: 0.87,
-      match_type: 'fts',
-      tags: ['数据结构'],
-    },
-    {
-      block_id: 'b3',
-      note_id: 'n-bst',
-      note_title: '二叉搜索树',
-      file_path: '/数据结构/二叉搜索树.md',
-      heading_path: '数据结构 / 二叉搜索树 / 基本操作',
-      snippet: '二叉搜索树的插入需要先找到合适的位置，再添加新节点...',
-      score: 0.72,
-      match_type: 'vector',
-      tags: ['数据结构', '树'],
-    },
-    {
-      block_id: 'b4',
-      note_id: 'n-deadlock',
-      note_title: '死锁',
-      file_path: '/操作系统/死锁.md',
-      heading_path: '操作系统 / 死锁 / 必要条件',
-      snippet: '死锁的四个必要条件：互斥、占有并等待、不可抢占、循环等待...',
-      score: 0.45,
-      match_type: 'vector',
-      tags: ['操作系统'],
-    },
-  ]
-  const filtered = results.filter(
-    (r) =>
-      r.note_title.includes(query) ||
-      r.snippet.includes(query) ||
-      r.heading_path.includes(query) ||
-      query.length > 1
-  )
-  return { results: filtered, total: filtered.length, mode }
-}

frontend/src/services/skillService.ts

@@ -42,77 +42,3 @@ export async function disableSkill(skillId: string): Promise<Skill> {
 export async function uninstallSkill(skillId: string): Promise<OperationResponse> {
   return apiClient.delete(`/api/skills/${skillId}`)
 }
-
-export const mockSkills: Skill[] = [
-  {
-    skill_id: 'exam-review',
-    name: '期末复习助手',
-    version: '1.0.0',
-    description: '根据课程笔记生成复习要点和练习题，帮助高效备考',
-    icon: '',
-    author: 'NotesAgent 团队',
-    permissions: ['notes.search', 'notes.read', 'tasks.create'],
-    tools: ['notes.search', 'notes.read', 'tasks.create'],
-    retrieval_config: { top_k: 10, rerank: true, citation: true },
-    model_requirements: { capabilities: ['chat', 'tool_calling'] },
-    status: 'ready',
-    enabled: true,
-  },
-  {
-    skill_id: 'meeting-summary',
-    name: '会议纪要生成',
-    version: '1.1.0',
-    description: '从音频或文本中提取会议要点、行动项和待办任务',
-    icon: '',
-    author: 'NotesAgent 团队',
-    permissions: ['notes.search', 'notes.write', 'tasks.write', 'attachments.read'],
-    tools: ['notes.search', 'notes.create', 'tasks.create', 'attachments.read'],
-    retrieval_config: { top_k: 5, rerank: false, citation: true },
-    model_requirements: { capabilities: ['chat', 'tool_calling', 'structured_output'] },
-    status: 'ready',
-    enabled: true,
-  },
-  {
-    skill_id: 'code-explainer',
-    name: '代码解读助手',
-    version: '0.9.0',
-    description: '分析代码片段，解释功能、复杂度和优化建议',
-    icon: '',
-    author: '社区贡献',
-    permissions: ['notes.search', 'notes.read'],
-    tools: ['notes.search', 'notes.read', 'rag.search'],
-    retrieval_config: { top_k: 8, rerank: true, citation: true },
-    model_requirements: { capabilities: ['chat', 'tool_calling'] },
-    status: 'installed',
-    enabled: false,
-  },
-  {
-    skill_id: 'research-assistant',
-    name: '文献研究助手',
-    version: '1.2.0',
-    description: '自动整理文献笔记，生成研究综述和引用关系图',
-    icon: '',
-    author: '社区贡献',
-    permissions: ['notes.search', 'notes.read', 'notes.write'],
-    tools: ['notes.search', 'notes.read', 'notes.create', 'rag.search'],
-    retrieval_config: { top_k: 15, rerank: true, citation: true },
-    model_requirements: { capabilities: ['chat', 'tool_calling', 'reasoning'] },
-    status: 'dependency_missing',
-    enabled: false,
-    missing_dependencies: ['文献引用插件', '知识图谱插件'],
-  },
-  {
-    skill_id: 'language-tutor',
-    name: '语言学习助手',
-    version: '0.5.0',
-    description: '基于你的学习笔记生成语言练习和记忆卡片',
-    icon: '',
-    author: '社区贡献',
-    permissions: ['notes.search', 'notes.read', 'tasks.create'],
-    tools: ['notes.search', 'notes.read', 'tasks.create'],
-    retrieval_config: { top_k: 6, rerank: false, citation: false },
-    model_requirements: { capabilities: ['chat'] },
-    status: 'ready',
-    enabled: true,
-  },
-]

frontend/src/services/systemService.ts

@@ -1,23 +1,14 @@
 import apiClient from './apiClient'
 import type { SystemStatus } from '@/contracts'
 
-export async function healthCheck(): Promise<{ status: string }> {
+export function healthCheck(): Promise<{ status: string }> {
-  try {
+  return apiClient.get('/health')
-    return await apiClient.get<{ status: string }>('/health')
-  } catch {
-    return { status: 'unavailable' }
-  }
 }
 
-export async function getStatus(): Promise<SystemStatus> {
+export function getStatus(): Promise<SystemStatus> {
-  try {
+  return apiClient.get('/api/status')
-    return await apiClient.get<SystemStatus>('/api/status')
+}
-  } catch {
+
-    return {
+export function getPermissionPolicy(): Promise<Record<string, 'allow' | 'confirm' | 'deny'>> {
-      status: 'ok',
+  return apiClient.get('/api/permissions/policy')
-      name: 'notes-agent',
-      version: '0.1.0',
-      environment: import.meta.env.DEV ? 'development' : 'production',
-    }
-  }
 }

frontend/src/services/taskService.ts

@@ -1,5 +1,5 @@
 import apiClient from './apiClient'
-import type { ApiTask, OperationResponse, PageMeta, TaskItem, TaskStatus, TaskPriority } from '@/contracts'
+import type { ApiTask, OperationResponse, PageMeta, TaskItem, TaskStatus } from '@/contracts'
 
 function toTask(task: ApiTask): TaskItem {
   return {
@@ -7,10 +7,8 @@ function toTask(task: ApiTask): TaskItem {
     title: task.title,
     description: task.description,
     status: task.status,
-    priority: 'medium',
     due_date: task.due_at ?? undefined,
     note_id: task.note_id ?? undefined,
-    source: 'user',
     created_at: task.created_at,
     updated_at: task.updated_at,
   }
@@ -60,67 +58,3 @@ export async function updateTask(
 export async function deleteTask(taskId: string): Promise<OperationResponse> {
   return apiClient.delete(`/api/tasks/${taskId}`)
 }
-
-export const mockTasks: TaskItem[] = [
-  {
-    task_id: 't-1',
-    title: '完成红黑树章节复习',
-    description: '整理插入、删除操作的所有情况，准备期末复习',
-    status: 'todo',
-    priority: 'high',
-    due_date: '2026-08-30T23:59:00Z',
-    note_id: 'n-rbt',
-    note_title: '红黑树',
-    source: 'user',
-    created_at: '2026-08-20T10:00:00Z',
-    updated_at: '2026-08-25T14:30:00Z',
-  },
-  {
-    task_id: 't-2',
-    title: '理解死锁的银行家算法',
-    description: '推导银行家算法的安全性检查过程',
-    status: 'in_progress',
-    priority: 'medium',
-    note_id: 'n-deadlock',
-    note_title: '死锁',
-    source: 'agent',
-    created_at: '2026-08-22T09:00:00Z',
-    updated_at: '2026-08-24T16:00:00Z',
-  },
-  {
-    task_id: 't-3',
-    title: 'TCP 三次握手与四次挥手',
-    description: '',
-    status: 'done',
-    priority: 'high',
-    note_id: 'n-tcp',
-    note_title: 'TCP_IP',
-    source: 'user',
-    created_at: '2026-08-15T08:00:00Z',
-    updated_at: '2026-08-18T20:00:00Z',
-  },
-  {
-    task_id: 't-4',
-    title: 'HTTP 状态码整理',
-    description: '整理常见 HTTP 状态码及含义',
-    status: 'todo',
-    priority: 'low',
-    note_id: 'n-http',
-    note_title: 'HTTP协议',
-    source: 'note',
-    created_at: '2026-08-10T10:00:00Z',
-    updated_at: '2026-08-10T10:00:00Z',
-  },
-  {
-    task_id: 't-5',
-    title: '链表操作实现练习',
-    description: '实现单链表和双向链表的基本操作',
-    status: 'todo',
-    priority: 'medium',
-    note_id: 'n-slist',
-    note_title: '单链表',
-    source: 'agent',
-    created_at: '2026-08-23T11:00:00Z',
-    updated_at: '2026-08-23T11:00:00Z',
-  },
-]

frontend/src/stores/agent.ts

@@ -1,21 +1,21 @@
 import { defineStore } from 'pinia'
 import { ref, computed } from 'vue'
 import type { AgentRun, AgentEvent, ToolDefinition, PermissionRequest, ToolCall } from '@/contracts'
-import { mockAgentRuns, mockAgentEvents, mockTools, mockPermissionRequest } from '@/services/agentService'
 import * as agentService from '@/services/agentService'
 import type { SseClient } from '@/services/sseClient'
 
 export const useAgentStore = defineStore('agent', () => {
-  const runs = ref<AgentRun[]>(mockAgentRuns)
+  const runs = ref<AgentRun[]>([])
-  const activeRunId = ref<string | null>('run-1')
+  const activeRunId = ref<string | null>(null)
-  const events = ref<AgentEvent[]>(mockAgentEvents.filter((e) => e.run_id === 'run-1'))
+  const events = ref<AgentEvent[]>([])
-  const tools = ref<ToolDefinition[]>(mockTools)
+  const tools = ref<ToolDefinition[]>([])
   const isCreating = ref(false)
   const isRunning = ref(false)
   const permissionRequest = ref<PermissionRequest | null>(null)
   const toolCalls = ref<ToolCall[]>([])
   const error = ref<string | null>(null)
   let eventStream: SseClient | null = null
+  let selectionVersion = 0
 
   const activeRun = computed(() =>
     runs.value.find((r) => r.run_id === activeRunId.value) || null
@@ -40,9 +40,15 @@ export const useAgentStore = defineStore('agent', () => {
   }
 
   async function loadRun(runId: string) {
+    const version = ++selectionVersion
     eventStream?.cancel()
     activeRunId.value = runId
+    events.value = []
+    toolCalls.value = []
+    permissionRequest.value = null
+    isRunning.value = false
     const run = await agentService.getAgentRun(runId)
+    if (version !== selectionVersion) return
     const existingIndex = runs.value.findIndex((item) => item.run_id === runId)
     if (existingIndex >= 0) runs.value[existingIndex] = run
     else runs.value.unshift(run)
@@ -106,9 +112,9 @@ export const useAgentStore = defineStore('agent', () => {
     isRunning.value = true
     error.value = null
     eventStream = agentService.streamAgentEvents(runId, {
-      onEvent: processEvent,
+      onEvent(event) { if (activeRunId.value === runId) processEvent(event) },
-      onError(streamError) { error.value = streamError.message; isRunning.value = false },
+      onError(streamError) { if (activeRunId.value === runId) { error.value = streamError.message; isRunning.value = false } },
-      onDone() { isRunning.value = false; eventStream = null },
+      onDone() { if (activeRunId.value === runId) { isRunning.value = false; eventStream = null } },
     })
   }
 
@@ -116,6 +122,7 @@ export const useAgentStore = defineStore('agent', () => {
     isCreating.value = true
     try {
       const run = await agentService.createAgentRun(request)
+      selectionVersion++
       runs.value.unshift(run)
       activeRunId.value = run.run_id
       events.value = []
@@ -143,10 +150,6 @@ export const useAgentStore = defineStore('agent', () => {
     permissionRequest.value = null
   }
 
-  function showPermissionDemo() {
-    permissionRequest.value = mockPermissionRequest
-  }
-
   return {
     runs,
     activeRunId,
@@ -166,6 +169,5 @@ export const useAgentStore = defineStore('agent', () => {
     createRun,
     cancelRun,
     respondPermission,
-    showPermissionDemo,
   }
 })

frontend/src/stores/chat.spec.ts

@@ -0,0 +1,43 @@
+import { beforeEach, expect, it, vi } from 'vitest'
+import { createPinia, setActivePinia } from 'pinia'
+import { useChatStore } from './chat'
+import { streamChat } from '@/services/chatService'
+import type { SseClient } from '@/services/sseClient'
+
+vi.mock('@/services/chatService', () => ({ streamChat: vi.fn() }))
+beforeEach(() => {
+  setActivePinia(createPinia())
+  vi.mocked(streamChat).mockReset().mockReturnValue({ cancel: vi.fn() } as unknown as SseClient)
+})
+
+it('sends real user history, applies streaming changes, and restores it when switching conversations', async () => {
+  const store = useChatStore()
+  store.selectedProviderId = 'real'
+  store.selectedModel = 'configured-model'
+  await store.sendMessage('user input')
+  const [request, handlers] = vi.mocked(streamChat).mock.calls[0]!
+  expect(request.messages).toEqual([{ role: 'user', content: 'user input' }])
+  handlers.onEvent?.({ event: 'TextDelta', sequence: 0, timestamp: '', data: { text: 'real response' } })
+  expect(store.messages[1]?.content).toBe('real response')
+  handlers.onDone?.()
+  const id = store.activeConversationId!
+  store.createNewConversation()
+  expect(store.messages).toEqual([])
+  await store.setActiveConversation(id)
+  expect(store.messages.map(m => m.content)).toEqual(['user input', 'real response'])
+})
+
+it('does not send without a provider and ignores late callbacks from a cancelled conversation', async () => {
+  const store = useChatStore()
+  await store.sendMessage('no provider')
+  expect(streamChat).not.toHaveBeenCalled()
+  store.selectedProviderId = 'real'
+  store.selectedModel = 'configured-model'
+  await store.sendMessage('first')
+  const old = vi.mocked(streamChat).mock.calls[0]![1]
+  store.createNewConversation()
+  await store.sendMessage('second')
+  old.onDone?.()
+  expect(store.isStreaming).toBe(true)
+  expect(store.messages[0]?.content).toBe('second')
+})