feat: add artifact library and improve public wiki

offline-backend-20260512/backend/CLAUDE.md

@@ -321,7 +321,7 @@ FastAPI application on port 8001 with health checks at `GET|HEAD /health`, `/api
 | **Uploads** (`/api/threads/{id}/uploads`) | `POST /` - upload files (auto-converts PDF/PPT/Excel/Word); `GET /list` - list; `DELETE /{filename}` - delete |
 | **Threads** (`/api/threads/{id}`) | `DELETE /` - remove DeerFlow-managed local thread data after LangGraph thread deletion; unexpected failures are logged server-side and return a generic 500 detail |
 | **Threads search** (`POST /threads/search`, LangGraph-compat) | Conversation listing. **System threads are excluded by default**: rows whose `metadata.system` is truthy (scheduler runs, roundtable workers) never crowd the user's list; a caller whose `metadata` filter explicitly references `thread_type`/`system` (e.g. studio notebooks) opts out of the exclusion. New `query` field = case-insensitive title (`display_name`) substring search, used by the chats page's server-side pagination. `POST /threads/count` takes the same request body (limit/offset ignored) and returns `{total}` under identical filtering/exclusion semantics ? powers the chats page's ??/??? display without fetching rows (`ThreadMetaStore.count`). `ThreadMetaStore.search` gained `exclude_system`/`query` params; JSON-blob filters batch-scan until the requested `limit`/`offset` window is full (so pagination over a heavily-filtered set never under-fills), with a `thread_id` tie-breaker on the `updated_at DESC` ordering for stable OFFSET paging. Tests: `tests/test_thread_meta_search.py` |
-| **Artifacts** (`/api/threads/{id}/artifacts`) | `GET /{path}` - serve artifacts; active content types (`text/html`, `application/xhtml+xml`, `image/svg+xml`) are always forced as download attachments to reduce XSS risk; `?download=true` still forces download for other file types |
+| **Artifacts** (`/api/threads/{id}/artifacts`, `/api/artifact-library`) | `GET /{path}` - serve artifacts; active content types (`text/html`, `application/xhtml+xml`, `image/svg+xml`) are always forced as download attachments to reduce XSS risk; `?download=true` still forces download for other file types. `GET /api/artifact-library` scans every current-user normal thread's authoritative `outputs/` directory (including files omitted from `present_files`), excludes uploads and image/audio/video media, joins the owning conversation title/agent route metadata, and exposes search/filter/pagination with a bounded 15-second per-user cache plus `refresh=true` bypass. Frontend route `/page/workspace/library` links each item back to the source chat with `?artifact=<path>` so the chat opens the existing artifact panel directly. |
 | **Word export** (`POST /api/writing/export/docx`) | Shared Markdown → DOCX generator used by conversation, artifact, scheduled-task, Deep Research, and AI-writing download surfaces. `app/gateway/word_export.py` builds OOXML with no Python conversion dependency: A4 portrait, 3.7/3.5/2.8/2.6 cm margins, real multilevel heading numbering, formal body/table styles, page numbers beginning at 1 with odd pages left-aligned and even pages right-aligned, and embedded `app/gateway/assets/fonts/方正小标宋简体.ttf`. The font loader validates its internal family, SHA-256, and OpenType embedding permission (`OS/2.fsType`) before export. Tests: `tests/test_word_export.py`. |
 | **Deep Research** (`/api/deep-research`) | Durable, user-owned research sessions/jobs with replayable SSE events and hidden-thread artifacts. `basic`/`quick` use the compact runner; `detailed` creates bounded subtopics, searches/writes sections concurrently, and assembles a report; `deep` recursively searches follow-up questions under a deterministic breadth/depth query budget. `multi_agent` is a LangGraph 1.x editor/researcher/writer/reviewer graph: its initial plan emits `awaiting_input`, the recorder stores the interrupt and releases the lease, and `POST /jobs/{id}/resume` rebuilds the graph from the persisted approved/revised plan. A bounded `custom_outline` is frozen into the session; headings/list entries deterministically supply the detailed and multi-agent plan, and the shared writer receives it only as a guarded content constraint. Completed reports support owner-scoped follow-up (`GET /sessions/{id}/messages`, `POST /sessions/{id}/chat`): context is only the current report, selected sources and recent messages; source-id citations are persisted, while `allow_new_research=true` is explicitly rejected. Owners may change that evidence set only after the job stops via `PATCH /sessions/{id}/sources/{source_id}`; this applies to future follow-ups rather than rewriting a completed report. Optional report illustrations are gated by `config.yaml -> deep_research.image_generation`: the Harness `OpenAICompatibleImageProvider` requires base64 output, the shared runner step writes only bounded `research-image-01..04.(png|jpg|webp)` artifacts and Markdown uses private `deep-research://` links. `GET /sessions/{id}/artifacts/{name}` checks session ownership before serving an image; an unavailable provider yields a warning but never fails the text report. Runners live in `packages/harness/deerflow/agents/deep_research/runners/`, may never import `app.*`, and use only the `AdapterBundle`. |
 | **Gateway auth** | Owner-scoped Deep Research routes await the platform's asynchronous `get_current_user` resolver before they read or write persistence, ensuring each session and job operation receives a concrete user ID. Deep Research artifacts are always resolved from a whitelisted filename via `/mnt/user-data/outputs/<name>` before mapping to a host path, preserving the sandbox guard on native Windows and Linux hosts. |
@@ -1527,8 +1527,10 @@ same DOM value, so a fetched-and-revoked object URL is not acceptable.
 Third-party pages embed published Wiki content through the chrome-free route
 `/#/embed/knowledge/{mapping_id}` (optional exact-page `wikiSlug`). It defaults
 to a recognized Wiki index page (title `索引`/`index`, page type `index`, or an
-`index`/`home`/`readme` slug), promotes that page to the first public-directory
-entry, then falls back to an overview/root page when no index exists. It retains
+`index`/`home`/`readme` slug) and promotes that page to the first public-directory
+entry. When upstream pages exist without an explicit index shell, the frontend
+injects a virtual `__index__` page and builds its catalogue from the page list;
+an all-empty dynamic index response must also fall back to that page list. It retains
 directory navigation and supports the shared public-embed `theme` / `bg` /
 `text` appearance inputs. WeKnora dynamically renders index page-type counts,
 nested category headings, page summaries, and `data-slug` links even though the
@@ -1544,9 +1546,14 @@ the loaded page title, slug, or alias and rewrites them to exact embedded
 `wikiSlug` routes. Preserve WeKnora's `wiki-content-link` interaction: primary
 brand color, medium weight, dashed underline becoming solid on hover, focus
 outline, and in-reader SPA navigation rather than a page reload. `GET
-/api/public/llmwiki/knowledge-bases` is the
-anonymous discovery endpoint and returns only each published ordinary mapping's
-`id` and `name` (plus the collection total). All anonymous reader calls use that
+/api/public/llmwiki/knowledge-bases` is the anonymous discovery endpoint. It
+validates published ordinary mappings against WeKnora, drops stale mappings,
+and returns safe comprehensive metadata from the WeKnora list/detail/page APIs:
+document/chunk/processing/share counts, non-secret configuration and timestamps,
+Wiki page type/status counts and explicit/generated index entry, plus aggregate
+local-vector state/counts/timestamps. Never expose remote WeKnora ids,
+tenant/creator identity, storage credentials, API credentials, raw vectors, or
+per-page indexing errors. All anonymous reader calls use that
 public namespace and the `/knowledge-bases/{id}` Wiki-page children. Those
 endpoints must resolve with a synthetic anonymous user and `is_admin=False`,
 explicitly require `publication_status=published`, and hard-deny the system

offline-backend-20260512/backend/README.md

@@ -153,6 +153,7 @@ FastAPI application providing REST endpoints for frontend integration:
 | `POST /api/threads/{id}/uploads` | Upload files (auto-converts PDF/PPT/Excel/Word to Markdown, rejects directory paths) |
 | `GET /api/threads/{id}/uploads/list` | List uploaded files |
 | `DELETE /api/threads/{id}` | Delete DeerFlow-managed local thread data after LangGraph thread deletion; unexpected failures are logged server-side and return a generic 500 detail |
+| `GET /api/artifact-library` | List the current user's generated conversation files across normal threads, with filename/conversation search, file-kind filtering, pagination, and an optional cache-bypassing refresh. The filesystem `outputs/` directories remain authoritative; uploads and image/audio/video files are excluded. |
 | `GET /api/threads/{id}/artifacts/{path}` | Serve generated artifacts |
 | `GET/POST /api/deep-research/*` | Durable Deep Research sessions and jobs. `quick`/`basic` run the compact report flow; `detailed` plans subtopics and writes independent sections; `deep` recursively investigates bounded follow-up branches. All modes reuse DeerFlow's configured material provider, LLM factory, hidden-thread artifacts, and SSE event log. Deep Research session/job/event/source/message writes serialise their rows in the writer transaction, rather than doing a post-commit ORM refresh or readback; this prevents MySQL read/write-splitting replica lag from failing “start writing”, dispatcher claim, report persistence, or follow-up history. Whole-document artifact/report rewrites likewise create their reversible version snapshots from writer-local ORM data and never refresh after commit, so replica lag cannot restore an otherwise successful rewrite. A report is written to the session and confirmed before its Job becomes `completed`; an unconfirmed write is retried and never exposed as a false completion. Retrying an already-created queued job re-nudges the dispatcher, recovering an interrupted HTTP response without creating a second report. If progress-event persistence, durable replay, or live fan-out is briefly unavailable, the worker/stream logs the degradation and continues or reconnects rather than failing the whole job. Whole-report rewrite accepts an optional `reportOutline` Markdown template; the durable job persists it and applies it as the chapter structure constraint. Optional report illustrations use a separately configured OpenAI-compatible image endpoint and are served only through the owning session's authenticated artifact route. |
 | `/api/enterprise-research/*` | Temporarily disabled. Its router and persistence initialization are not registered, so these endpoints return 404 and no enterprise-research tables are created or migrated during startup. |
@@ -865,8 +866,10 @@ report-structure column for databases stamped by either old variant.
 Published knowledge bases also have a login-free, chrome-free reader at
 `/#/embed/knowledge/{mapping_id}` for direct third-party iframe use. With no
 `wikiSlug`, it selects the Wiki **索引** page first (recognized by title, page
-type, or `index`/`home`/`readme` slug), pins that page to the top of the public
-directory, and otherwise falls back to an overview/root page;
+type, or `index`/`home`/`readme` slug) and pins that page to the top of the
+public directory. If WeKnora has pages but no explicit index page, the reader
+creates a virtual `__index__` landing page from the public directory; empty
+dynamic-index groups also fall back to those page summaries;
 `?wikiSlug=...` deep-links one page. The reader supports
 `theme=light|dark`, `bg`, and `text` embed appearance parameters. Index-article
 catalogue sections (page-type counts, nested categories, page summaries) are
@@ -882,8 +885,14 @@ only from the read-only `/api/public/llmwiki/knowledge-bases/{id}/...` namespace
 which rechecks `publication_status=published` on every request and denies the
 system conversation-deposit knowledge base. Private or unpublished bases are
 never exposed by this route. Anonymous consumers can discover the available
-public Wiki libraries through `GET /api/public/llmwiki/knowledge-bases`; each
-entry deliberately contains only the DeerFlow mapping `id` and display `name`.
+public Wiki libraries through `GET /api/public/llmwiki/knowledge-bases`. The
+endpoint validates each published mapping against WeKnora, omits stale remote
+mappings, and returns safe comprehensive metadata: description/type,
+document/chunk/processing/share counts, non-secret configuration and
+timestamps, Wiki page/type/status counts plus explicit/generated index mode and
+default embed route, and aggregate local-vector state/counts/timestamps. It
+never returns the WeKnora id, tenant/creator identity, storage credentials, API
+credentials, raw vectors, or per-page indexing errors.
 Protected `resource://` and supported object-storage images are fetched through
 the current mapping instead of being exposed as upstream URLs. Authenticated
 readers use `GET /api/llmwiki/knowledge-bases/{id}/files?file_path=...`; public

offline-backend-20260512/backend/app/gateway/routers/artifacts.py

@@ -1,11 +1,15 @@
+import asyncio
 import logging
 import mimetypes
+import time
 import zipfile
-from datetime import datetime, timezone
+from collections import OrderedDict
+from datetime import UTC, datetime
 from pathlib import Path
+from typing import Literal
 from urllib.parse import quote
 
-from fastapi import APIRouter, HTTPException, Request
+from fastapi import APIRouter, HTTPException, Query, Request
 from fastapi.responses import FileResponse, PlainTextResponse, Response
 from pydantic import BaseModel
 
@@ -13,6 +17,7 @@ from app.gateway.authz import require_permission
 from app.gateway.path_utils import aresolve_thread_virtual_path
 from deerflow.config.paths import get_paths
 from deerflow.runtime.thread_paths import aresolve_path_user_id
+from deerflow.runtime.user_context import get_effective_user_id
 
 logger = logging.getLogger(__name__)
 
@@ -38,6 +43,84 @@ EDITABLE_TEXT_SUFFIXES = {
 
 _VIRTUAL_USER_DATA_PREFIX = "mnt/user-data"
 
+ArtifactLibraryKind = Literal["all", "document", "spreadsheet", "text_data", "archive", "other"]
+
+_MEDIA_MIME_PREFIXES = ("image/", "audio/", "video/")
+_MEDIA_SUFFIXES = {
+    ".aac",
+    ".aiff",
+    ".ape",
+    ".avi",
+    ".bmp",
+    ".flac",
+    ".gif",
+    ".heic",
+    ".ico",
+    ".jpeg",
+    ".jpg",
+    ".m4a",
+    ".m4v",
+    ".mkv",
+    ".mov",
+    ".mp3",
+    ".mp4",
+    ".ogg",
+    ".png",
+    ".svg",
+    ".tiff",
+    ".wav",
+    ".webm",
+    ".webp",
+    ".wma",
+}
+_DOCUMENT_SUFFIXES = {
+    ".doc",
+    ".docx",
+    ".epub",
+    ".odt",
+    ".pdf",
+    ".ppt",
+    ".pptx",
+    ".rtf",
+}
+_SPREADSHEET_SUFFIXES = {".csv", ".ods", ".tsv", ".xls", ".xlsx"}
+_TEXT_DATA_SUFFIXES = {
+    ".bash",
+    ".c",
+    ".conf",
+    ".cpp",
+    ".css",
+    ".go",
+    ".h",
+    ".html",
+    ".ini",
+    ".java",
+    ".js",
+    ".json",
+    ".jsonl",
+    ".log",
+    ".markdown",
+    ".md",
+    ".mmd",
+    ".py",
+    ".rs",
+    ".sh",
+    ".sql",
+    ".toml",
+    ".ts",
+    ".tsx",
+    ".txt",
+    ".xml",
+    ".yaml",
+    ".yml",
+}
+_ARCHIVE_SUFFIXES = {".7z", ".gz", ".rar", ".skill", ".tar", ".tgz", ".zip"}
+_ARTIFACT_LIBRARY_CACHE_TTL_SECONDS = 15.0
+_ARTIFACT_LIBRARY_CACHE_MAX_USERS = 128
+_ARTIFACT_LIBRARY_SCAN_CONCURRENCY = 8
+_ARTIFACT_LIBRARY_THREAD_BATCH_SIZE = 500
+_artifact_library_cache: OrderedDict[str, tuple[float, list[dict]]] = OrderedDict()
+
 
 def coerce_artifact_virtual_path(path: str) -> str:
     """Normalize an artifacts URL path to a sandbox virtual path.
@@ -60,6 +143,26 @@ def coerce_artifact_virtual_path(path: str) -> str:
 
 class ArtifactUpdateRequest(BaseModel):
     content: str
+
+
+class ArtifactLibraryItem(BaseModel):
+    thread_id: str
+    thread_title: str
+    agent_id: str | None = None
+    thread_type: str | None = None
+    path: str
+    name: str
+    size_bytes: int
+    mime_type: str | None = None
+    modified_at: str
+    file_kind: str
+
+
+class ArtifactLibraryResponse(BaseModel):
+    total: int
+    offset: int
+    limit: int
+    items: list[ArtifactLibraryItem]
 
 
 def _build_content_disposition(disposition_type: str, filename: str) -> str:
@@ -139,21 +242,184 @@ def collect_outputs_listing(outputs_dir: Path) -> list[dict]:
             resolved.relative_to(outputs_root)
         except (OSError, ValueError):
             continue
-        relative = resolved.relative_to(outputs_root).as_posix()
-        mime_type, _ = mimetypes.guess_type(resolved.name)
-        stat = resolved.stat()
+        try:
+            relative = resolved.relative_to(outputs_root).as_posix()
+            mime_type, _ = mimetypes.guess_type(resolved.name)
+            stat = resolved.stat()
+        except OSError:
+            # A run may atomically replace or remove a file while this directory
+            # is being scanned. Treat that transient entry as absent instead of
+            # failing the complete thread/library listing.
+            continue
         files.append(
             {
                 "path": f"mnt/user-data/outputs/{relative}",
                 "name": resolved.name,
                 "size_bytes": stat.st_size,
                 "mime_type": mime_type,
-                "modified_at": datetime.fromtimestamp(stat.st_mtime, tz=timezone.utc).isoformat(),
+                "modified_at": datetime.fromtimestamp(stat.st_mtime, tz=UTC).isoformat(),
             }
         )
 
     files.sort(key=lambda f: f["modified_at"], reverse=True)
     return files
+
+
+def artifact_library_kind(name: str) -> str:
+    """Classify one non-media artifact for the library filters."""
+    suffix = Path(name).suffix.lower()
+    if suffix in _DOCUMENT_SUFFIXES:
+        return "document"
+    if suffix in _SPREADSHEET_SUFFIXES:
+        return "spreadsheet"
+    if suffix in _TEXT_DATA_SUFFIXES:
+        return "text_data"
+    if suffix in _ARCHIVE_SUFFIXES:
+        return "archive"
+    return "other"
+
+
+def is_artifact_library_file(file: dict) -> bool:
+    """Return whether a generated file belongs in the document-only library."""
+    mime_type = str(file.get("mime_type") or "").lower()
+    if mime_type.startswith(_MEDIA_MIME_PREFIXES):
+        return False
+    return Path(str(file.get("name") or file.get("path") or "")).suffix.lower() not in _MEDIA_SUFFIXES
+
+
+async def _search_artifact_library_threads(thread_store) -> list[dict]:
+    """Load every normal conversation owned by the current request user."""
+    rows: list[dict] = []
+    seen: set[str] = set()
+    offset = 0
+    while True:
+        batch = await thread_store.search(
+            limit=_ARTIFACT_LIBRARY_THREAD_BATCH_SIZE,
+            offset=offset,
+            exclude_system=True,
+        )
+        if not batch:
+            break
+        added = 0
+        for row in batch:
+            thread_id = str(row.get("thread_id") or "")
+            if not thread_id or thread_id in seen:
+                continue
+            seen.add(thread_id)
+            rows.append(row)
+            added += 1
+        if len(batch) < _ARTIFACT_LIBRARY_THREAD_BATCH_SIZE or added == 0:
+            break
+        offset += len(batch)
+    return rows
+
+
+async def _scan_artifact_library(thread_store) -> list[dict]:
+    """Scan generated outputs for all current-user conversation threads."""
+    rows = await _search_artifact_library_threads(thread_store)
+    paths = get_paths()
+    semaphore = asyncio.Semaphore(_ARTIFACT_LIBRARY_SCAN_CONCURRENCY)
+
+    async def scan_thread(row: dict) -> list[dict]:
+        thread_id = str(row.get("thread_id") or "")
+        if not thread_id:
+            return []
+        async with semaphore:
+            try:
+                path_user_id = await aresolve_path_user_id(thread_id)
+                outputs_dir = paths.sandbox_outputs_dir(thread_id, user_id=path_user_id)
+                files = await asyncio.to_thread(collect_outputs_listing, outputs_dir)
+            except Exception:
+                logger.warning("Failed to scan artifacts for library thread %s", thread_id, exc_info=True)
+                return []
+
+        metadata = row.get("metadata") if isinstance(row.get("metadata"), dict) else {}
+        agent_id = metadata.get("agent_id")
+        thread_type = metadata.get("thread_type")
+        title = str(row.get("display_name") or "新对话")
+        items: list[dict] = []
+        for file in files:
+            if not is_artifact_library_file(file):
+                continue
+            items.append(
+                {
+                    "thread_id": thread_id,
+                    "thread_title": title,
+                    "agent_id": agent_id if isinstance(agent_id, str) and agent_id else None,
+                    "thread_type": thread_type if isinstance(thread_type, str) and thread_type else None,
+                    **file,
+                    "file_kind": artifact_library_kind(str(file.get("name") or "")),
+                }
+            )
+        return items
+
+    nested = await asyncio.gather(*(scan_thread(row) for row in rows))
+    items = [item for group in nested for item in group]
+    items.sort(
+        key=lambda item: (
+            str(item.get("modified_at") or ""),
+            str(item.get("thread_id") or ""),
+            str(item.get("path") or ""),
+        ),
+        reverse=True,
+    )
+    return items
+
+
+async def _cached_artifact_library(thread_store, *, refresh: bool) -> list[dict]:
+    user_id = str(get_effective_user_id() or "default")
+    now = time.monotonic()
+    cached = _artifact_library_cache.get(user_id)
+    if not refresh and cached and cached[0] > now:
+        _artifact_library_cache.move_to_end(user_id)
+        return cached[1]
+
+    items = await _scan_artifact_library(thread_store)
+    _artifact_library_cache[user_id] = (now + _ARTIFACT_LIBRARY_CACHE_TTL_SECONDS, items)
+    _artifact_library_cache.move_to_end(user_id)
+    while len(_artifact_library_cache) > _ARTIFACT_LIBRARY_CACHE_MAX_USERS:
+        _artifact_library_cache.popitem(last=False)
+    return items
+
+
+@router.get(
+    "/artifact-library",
+    response_model=ArtifactLibraryResponse,
+    summary="List Generated Files Across Conversations",
+    description="Aggregate non-media files generated in the current user's normal conversations.",
+)
+@require_permission("threads", "read")
+async def list_artifact_library(
+    request: Request,
+    query: str | None = Query(default=None, max_length=200),
+    file_kind: ArtifactLibraryKind = Query(default="all"),
+    offset: int = Query(default=0, ge=0),
+    limit: int = Query(default=50, ge=1, le=200),
+    refresh: bool = Query(default=False),
+) -> ArtifactLibraryResponse:
+    """Return a flat, newest-first library of generated conversation files.
+
+    The filesystem remains the source of truth: every owned, non-system thread's
+    ``outputs`` directory is scanned, including files written without an
+    explicit ``present_files`` call. Uploads and image/audio/video outputs are
+    intentionally outside this collection.
+    """
+    from app.gateway.deps import get_thread_store
+
+    items = await _cached_artifact_library(get_thread_store(request), refresh=refresh)
+    normalized_query = (query or "").strip().casefold()
+    filtered = [
+        item
+        for item in items
+        if (file_kind == "all" or item["file_kind"] == file_kind) and (not normalized_query or normalized_query in str(item.get("name") or "").casefold() or normalized_query in str(item.get("thread_title") or "").casefold())
+    ]
+    page = filtered[offset : offset + limit]
+    return ArtifactLibraryResponse(
+        total=len(filtered),
+        offset=offset,
+        limit=limit,
+        items=[ArtifactLibraryItem.model_validate(item) for item in page],
+    )
 
 
 @router.get(

offline-backend-20260512/backend/app/gateway/routers/llmwiki.py

@@ -55,6 +55,7 @@ from deerflow.config.agents_config import load_agent_config
 from deerflow.config.llmwiki_config import LlmWikiRuntimeOverride, normalize_service_url
 from deerflow.config.system_settings import load_system_settings, save_system_settings
 from deerflow.integrations.weknora.client import WeKnoraClient, WeKnoraError
+from deerflow.integrations.weknora.local_index.retrieval import is_fully_vectorized
 from deerflow.integrations.weknora.local_index.search import WikiIndexError
 from deerflow.integrations.weknora.runtime import build_weknora_client, get_resolved_llmwiki_runtime
 from deerflow.persistence.llmwiki import LlmWikiStore
@@ -496,15 +497,7 @@ def _should_follow_weknora_binary_redirect(path: str) -> bool:
     """Keep protected preview/image bytes inside the DeerFlow proxy."""
 
     segments = _path_segments(path)
-    return (
-        len(segments) == 5
-        and segments[:3] == ["api", "v1", "knowledge"]
-        and segments[4] == "preview"
-    ) or (
-        len(segments) == 5
-        and segments[:3] == ["api", "v1", "knowledge-bases"]
-        and segments[4] == "files"
-    )
+    return (len(segments) == 5 and segments[:3] == ["api", "v1", "knowledge"] and segments[4] == "preview") or (len(segments) == 5 and segments[:3] == ["api", "v1", "knowledge-bases"] and segments[4] == "files")
 
 
 def _forward_request_headers(request: Request) -> dict[str, str]:
@@ -1833,6 +1826,9 @@ def _mapping_view(row: dict[str, Any], remote: dict[str, Any] | None, *, actor_u
         "knowledge_count": int(source.get("knowledge_count") or 0),
         "chunk_count": int(source.get("chunk_count") or 0),
         "processing_count": int(source.get("processing_count") or 0),
+        "share_count": int(source.get("share_count") or 0),
+        "is_processing": bool(source.get("is_processing") or int(source.get("processing_count") or 0) > 0),
+        "is_temporary": bool(source.get("is_temporary", False)),
         "embedding_model_id": source.get("embedding_model_id"),
         "vector_store_id": source.get("vector_store_id"),
         "vector_store_name": source.get("vector_store_name"),
@@ -1840,16 +1836,23 @@ def _mapping_view(row: dict[str, Any], remote: dict[str, Any] | None, *, actor_u
         "vector_store_engine_type": source.get("vector_store_engine_type"),
         "vector_store_status": source.get("vector_store_status"),
         "chunking_config": source.get("chunking_config") if isinstance(source.get("chunking_config"), dict) else {},
+        "image_processing_config": (source.get("image_processing_config") if isinstance(source.get("image_processing_config"), dict) else {}),
         "indexing_strategy": source.get("indexing_strategy") if isinstance(source.get("indexing_strategy"), dict) else {},
         "capabilities": source.get("capabilities") if isinstance(source.get("capabilities"), dict) else {},
         "extract_config": source.get("extract_config") if isinstance(source.get("extract_config"), dict) else {},
+        "faq_config": source.get("faq_config") if isinstance(source.get("faq_config"), dict) else {},
         "wiki_config": source.get("wiki_config") if isinstance(source.get("wiki_config"), dict) else {},
         "question_generation_config": (source.get("question_generation_config") if isinstance(source.get("question_generation_config"), dict) else {}),
+        "auto_tag_config": source.get("auto_tag_config") if isinstance(source.get("auto_tag_config"), dict) else {},
+        "vlm_config": source.get("vlm_config") if isinstance(source.get("vlm_config"), dict) else {},
+        "asr_config": source.get("asr_config") if isinstance(source.get("asr_config"), dict) else {},
         "summary_model_id": source.get("summary_model_id"),
         "storage_provider": (source.get("storage_provider_config", {}).get("provider") if isinstance(source.get("storage_provider_config"), dict) else None),
         "created_at": row.get("created_at"),
         "updated_at": row.get("updated_at"),
         "published_at": row.get("published_at"),
+        "remote_created_at": source.get("created_at"),
+        "remote_updated_at": source.get("updated_at"),
     }
 
 
@@ -1877,11 +1880,7 @@ async def _public_mapping(request: Request, mapping_id: str) -> dict[str, Any]:
         write=False,
         is_admin=False,
     )
-    if (
-        row is None
-        or row.get("publication_status") != "published"
-        or _is_conversation_deposit_mapping(row)
-    ):
+    if row is None or row.get("publication_status") != "published" or _is_conversation_deposit_mapping(row):
         raise HTTPException(status_code=404, detail="Published knowledge base not found")
     return row
 
@@ -2537,24 +2536,260 @@ async def test_llmwiki_connection(request: Request, body: LlmWikiConnectionTest)
     return {"success": True, "provider": "weknora", "health": health}
 
 
+_PUBLIC_WIKI_SUMMARY_CONCURRENCY = 8
+_PUBLIC_WIKI_PAGE_SIZE = 500
+_PUBLIC_WIKI_PAGE_LIMIT = 100
+_GENERATED_PUBLIC_WIKI_INDEX_SLUG = "__index__"
+
+
+def _is_public_wiki_index_page(page: dict[str, Any]) -> bool:
+    page_type = str(page.get("page_type") or "").strip().lower()
+    title = str(page.get("title") or "").strip().lower()
+    slug_name = str(page.get("slug") or "").strip("/").rsplit("/", 1)[-1].lower()
+    return page_type == "index" or title in {"索引", "index"} or slug_name in {"index", "home", "readme"}
+
+
+def _public_wiki_default_page(pages: list[dict[str, Any]]) -> dict[str, Any] | None:
+    return (
+        next((page for page in pages if _is_public_wiki_index_page(page)), None)
+        or next((page for page in pages if str(page.get("page_type") or "").lower() == "overview"), None)
+        or next(
+            (page for page in pages if not str(page.get("parent_slug") or "").strip() and int(page.get("depth") or 0) == 0),
+            None,
+        )
+        or (pages[0] if pages else None)
+    )
+
+
+def _count_public_wiki_values(pages: list[dict[str, Any]], key: str, fallback: str) -> dict[str, int]:
+    counts: dict[str, int] = {}
+    for page in pages:
+        value = str(page.get(key) or fallback).strip().lower() or fallback
+        counts[value] = counts.get(value, 0) + 1
+    return counts
+
+
+def _public_wiki_time_bound(pages: list[dict[str, Any]], key: str, *, latest: bool) -> str | None:
+    values = [str(page.get(key) or "").strip() for page in pages]
+    values = [value for value in values if value]
+    if not values:
+        return None
+    return max(values) if latest else min(values)
+
+
+async def _all_public_wiki_page_views(
+    client: WeKnoraClient,
+    remote_id: str,
+) -> tuple[list[dict[str, Any]], int, bool]:
+    pages_by_slug: dict[str, dict[str, Any]] = {}
+    remote_total = 0
+    complete = False
+    for page_number in range(1, _PUBLIC_WIKI_PAGE_LIMIT + 1):
+        result = await client.list_wiki_pages(
+            remote_id,
+            page=page_number,
+            page_size=_PUBLIC_WIKI_PAGE_SIZE,
+            query="",
+        )
+        raw_pages = result.get("pages") if isinstance(result, dict) else []
+        page_items = raw_pages if isinstance(raw_pages, list) else []
+        for page_item in page_items:
+            if not isinstance(page_item, dict):
+                continue
+            page = _wiki_page_view(page_item)
+            if page is None:
+                continue
+            slug = str(page.get("slug") or "").strip()
+            pages_by_slug[slug or f"__page_{len(pages_by_slug)}"] = page
+        reported_total = int(result.get("total") or 0)
+        remote_total = max(remote_total, reported_total, len(pages_by_slug))
+        total_pages = max(1, int(result.get("total_pages") or 1))
+        if not page_items or page_number >= total_pages or (reported_total > 0 and len(pages_by_slug) >= remote_total):
+            complete = True
+            break
+    return list(pages_by_slug.values()), max(remote_total, len(pages_by_slug)), complete
+
+
+def _public_wiki_summary(
+    mapping_id: str,
+    pages: list[dict[str, Any]],
+    *,
+    total: int,
+    complete: bool,
+) -> dict[str, Any]:
+    explicit_index = next((page for page in pages if _is_public_wiki_index_page(page)), None)
+    default_page = _public_wiki_default_page(pages)
+    if explicit_index is not None:
+        index_mode = "explicit"
+        index_slug: str | None = str(explicit_index.get("slug") or "") or None
+    elif pages:
+        index_mode = "generated"
+        index_slug = _GENERATED_PUBLIC_WIKI_INDEX_SLUG
+    else:
+        index_mode = "none"
+        index_slug = None
+    default_slug = index_slug or (str(default_page.get("slug") or "") if default_page else None)
+    route = f"/embed/knowledge/{quote(mapping_id, safe='')}"
+    if default_slug:
+        route = f"{route}?{urlencode({'wikiSlug': default_slug})}"
+    return {
+        "status": "ready" if pages else "empty",
+        "page_count": total,
+        "loaded_page_count": len(pages),
+        "stats_complete": complete,
+        "page_type_counts": _count_public_wiki_values(pages, "page_type", "page"),
+        "page_status_counts": _count_public_wiki_values(pages, "status", "unknown"),
+        "index_mode": index_mode,
+        "has_explicit_index": explicit_index is not None,
+        "index_slug": index_slug,
+        "default_slug": default_slug,
+        "created_at": _public_wiki_time_bound(pages, "created_at", latest=False),
+        "updated_at": _public_wiki_time_bound(pages, "updated_at", latest=True),
+        "embed_route": route,
+    }
+
+
+def _unavailable_public_wiki_summary(mapping_id: str) -> dict[str, Any]:
+    return {
+        "status": "unavailable",
+        "page_count": 0,
+        "loaded_page_count": 0,
+        "stats_complete": False,
+        "page_type_counts": {},
+        "page_status_counts": {},
+        "index_mode": "unknown",
+        "has_explicit_index": False,
+        "index_slug": None,
+        "default_slug": None,
+        "created_at": None,
+        "updated_at": None,
+        "embed_route": f"/embed/knowledge/{quote(mapping_id, safe='')}",
+    }
+
+
+def _public_local_index_summary(
+    request: Request,
+    mapping: dict[str, Any],
+    state: dict[str, Any] | None,
+) -> dict[str, Any]:
+    app_state = request.app.state
+    local_config = getattr(
+        getattr(getattr(app_state, "config", None), "llmwiki", None),
+        "local_wiki_index",
+        None,
+    )
+    enabled = bool(getattr(local_config, "enabled", False))
+    current = dict(state or {})
+    state_name = str(current.get("state") or "not_synced")
+    if state_name == "idle" and int(current.get("index_revision") or 0) == 0 and not current.get("last_completed_at") and not current.get("last_success_at"):
+        state_name = "not_synced"
+    embedding = getattr(app_state, "llmwiki_embedding", None)
+    fingerprint = embedding.fingerprint if embedding is not None else None
+    fully_vectorized = bool(mapping.get("wiki_index_enabled", True)) and is_fully_vectorized(
+        current,
+        fingerprint,
+    )
+    return {
+        "enabled": enabled,
+        "wiki_index_enabled": bool(mapping.get("wiki_index_enabled", True)),
+        "state": state_name,
+        "fully_vectorized": fully_vectorized,
+        "retrieval_mode": "vector" if fully_vectorized else "wiki_api",
+        "remote_page_count": int(current.get("remote_page_count") or 0),
+        "local_page_count": int(current.get("local_page_count") or 0),
+        "ready_page_count": int(current.get("ready_page_count") or 0),
+        "failed_page_count": int(current.get("failed_page_count") or 0),
+        "vector_count": int(current.get("vector_count") or 0),
+        "last_started_at": current.get("last_started_at"),
+        "last_completed_at": current.get("last_completed_at"),
+        "last_success_at": current.get("last_success_at"),
+        "updated_at": current.get("updated_at"),
+    }
+
+
 @router.get("/api/public/llmwiki/knowledge-bases")
 async def list_public_knowledge_bases(request: Request) -> dict[str, Any]:
-    """List the minimal identity of every knowledge base published as a public Wiki."""
+    """List live, published Wiki libraries with safe metadata and index health."""
 
-    rows = await _store(request).list_visible(
-        "__public_wiki__",
-        scope="public",
-        is_admin=False,
-    )
-    knowledge_bases = [
-        {
-            "id": str(row["id"]),
-            "name": str(row.get("name") or ""),
-        }
-        for row in rows
-        if row.get("publication_status") == "published"
-        and not _is_conversation_deposit_mapping(row)
+    rows = [
+        row
+        for row in await _store(request).list_visible(
+            "__public_wiki__",
+            scope="public",
+            is_admin=False,
+        )
+        if row.get("publication_status") == "published" and not _is_conversation_deposit_mapping(row)
     ]
+    client = _client_or_503()
+    try:
+        remote_items = await client.list_knowledge_bases()
+    except WeKnoraError as exc:
+        _raise_upstream(exc)
+    remote_by_id = {str(item.get("id")): item for item in remote_items if isinstance(item, dict) and item.get("id")}
+    # A successful upstream list is authoritative: stale DeerFlow mappings are
+    # omitted so discovery never advertises a Wiki that its detail route cannot open.
+    rows = [row for row in rows if str(row.get("weknora_id") or "") in remote_by_id]
+    mapping_ids = [str(row["id"]) for row in rows]
+    index_states: dict[str, dict[str, Any]] = {}
+    index_store = getattr(request.app.state, "llmwiki_index_store", None)
+    if index_store is not None and mapping_ids:
+        try:
+            index_states = {str(state.get("knowledge_base_mapping_id") or ""): state for state in await index_store.list_index_status(mapping_ids) if isinstance(state, dict)}
+        except Exception:
+            logger.warning("Failed to read public Wiki local-index summaries", exc_info=True)
+
+    semaphore = asyncio.Semaphore(_PUBLIC_WIKI_SUMMARY_CONCURRENCY)
+
+    async def build_summary(row: dict[str, Any]) -> dict[str, Any]:
+        mapping_id = str(row["id"])
+        remote_id = str(row["weknora_id"])
+        remote = remote_by_id[remote_id]
+        async with semaphore:
+            try:
+                remote = {**remote, **await client.get_knowledge_base(remote_id)}
+            except WeKnoraError as exc:
+                if exc.status_code == 404:
+                    return {}
+                logger.warning(
+                    "Failed to load public Wiki detail metadata: mapping_id=%s",
+                    mapping_id,
+                    exc_info=True,
+                )
+            try:
+                pages, page_total, complete = await _all_public_wiki_page_views(client, remote_id)
+                wiki = _public_wiki_summary(
+                    mapping_id,
+                    pages,
+                    total=page_total,
+                    complete=complete,
+                )
+            except WeKnoraError as exc:
+                logger.warning(
+                    "Failed to load public Wiki page summary: mapping_id=%s",
+                    mapping_id,
+                    exc_info=True,
+                )
+                wiki = (
+                    _public_wiki_summary(mapping_id, [], total=0, complete=True)
+                    if exc.status_code == 404
+                    else _unavailable_public_wiki_summary(mapping_id)
+                )
+        return {
+            **_mapping_view(
+                row,
+                remote,
+                actor_user_id="__public_wiki__",
+                is_admin=False,
+            ),
+            "wiki": wiki,
+            "local_index": _public_local_index_summary(
+                request,
+                row,
+                index_states.get(mapping_id),
+            ),
+        }
+
+    knowledge_bases = [summary for summary in await asyncio.gather(*(build_summary(row) for row in rows)) if summary]
     return {"knowledge_bases": knowledge_bases, "total": len(knowledge_bases)}
 
 
@@ -2596,13 +2831,7 @@ async def list_public_wiki_pages(
     except WeKnoraError as exc:
         _raise_upstream(exc)
     raw_pages = result.get("pages") if isinstance(result, dict) else []
-    pages = [
-        view
-        for page_item in (raw_pages if isinstance(raw_pages, list) else [])
-        if isinstance(page_item, dict)
-        for view in [_wiki_page_view(page_item)]
-        if view is not None
-    ]
+    pages = [view for page_item in (raw_pages if isinstance(raw_pages, list) else []) if isinstance(page_item, dict) for view in [_wiki_page_view(page_item)] if view is not None]
     return {**result, "pages": pages}
 
 
@@ -2650,11 +2879,7 @@ def _public_wiki_index_item(value: Any) -> dict[str, Any] | None:
         "slug": slug,
         "title": str(value.get("title") or slug),
         "summary": str(value.get("summary") or ""),
-        "category_path": (
-            [str(item) for item in value.get("category_path", []) if str(item).strip()]
-            if isinstance(value.get("category_path"), list)
-            else []
-        ),
+        "category_path": ([str(item) for item in value.get("category_path", []) if str(item).strip()] if isinstance(value.get("category_path"), list) else []),
         "wiki_path": str(value.get("wiki_path") or ""),
         "depth": int(value.get("depth") or 0),
     }
@@ -2680,20 +2905,12 @@ async def _all_public_wiki_index(client: WeKnoraClient, remote_id: str) -> dict[
             version = max(version, int(payload.get("version") or 0))
             raw_groups = payload.get("groups") if isinstance(payload.get("groups"), list) else []
             group = next(
-                (
-                    value
-                    for value in raw_groups
-                    if isinstance(value, dict) and value.get("type") == page_type
-                ),
+                (value for value in raw_groups if isinstance(value, dict) and value.get("type") == page_type),
                 {},
             )
             total = max(total, int(group.get("total") or 0))
             raw_items = group.get("items") if isinstance(group.get("items"), list) else []
-            items.extend(
-                item
-                for value in raw_items
-                if (item := _public_wiki_index_item(value)) is not None
-            )
+            items.extend(item for value in raw_items if (item := _public_wiki_index_item(value)) is not None)
             next_cursor = str(group.get("next_cursor") or "")
             if not next_cursor or next_cursor in seen_cursors:
                 break
@@ -3201,12 +3418,7 @@ async def search_llmwiki(request: Request, body: SearchRequest) -> dict[str, Any
     except WikiIndexError as exc:
         raise HTTPException(status_code=exc.status_code, detail={"code": exc.code, "message": str(exc)}) from None
     mapping_by_id = {str(row["id"]): row for row in rows}
-    result["results"] = [
-        _sanitize_conversation_deposit_value(item)
-        if _is_conversation_deposit_mapping(mapping_by_id.get(str(item.get("knowledge_base_id") or ""), {}))
-        else item
-        for item in result.get("results") or []
-    ]
+    result["results"] = [_sanitize_conversation_deposit_value(item) if _is_conversation_deposit_mapping(mapping_by_id.get(str(item.get("knowledge_base_id") or ""), {})) else item for item in result.get("results") or []]
     return result
 
 

offline-backend-20260512/backend/tests/test_artifact_library.py

@@ -0,0 +1,142 @@
+from __future__ import annotations
+
+import os
+from pathlib import Path
+from types import SimpleNamespace
+
+import pytest
+
+from app.gateway import deps
+from app.gateway.routers import artifacts
+
+
+class _FakeThreadStore:
+    def __init__(self, rows: list[dict]) -> None:
+        self.rows = rows
+        self.calls: list[dict] = []
+
+    async def search(self, **kwargs) -> list[dict]:
+        self.calls.append(kwargs)
+        offset = int(kwargs.get("offset", 0))
+        limit = int(kwargs.get("limit", 100))
+        return self.rows[offset : offset + limit]
+
+
+@pytest.mark.asyncio
+async def test_artifact_library_aggregates_generated_non_media_files(tmp_path: Path, monkeypatch):
+    thread_one = tmp_path / "thread-one" / "outputs"
+    thread_two = tmp_path / "thread-two" / "outputs"
+    thread_one.mkdir(parents=True)
+    thread_two.mkdir(parents=True)
+
+    report = thread_one / "研判报告.docx"
+    table = thread_one / "数据汇总.xlsx"
+    image = thread_one / "配图.png"
+    audio = thread_two / "录音.mp3"
+    notes = thread_two / "结论.md"
+    archive = thread_two / "证据.zip"
+    for path, content in (
+        (report, b"doc"),
+        (table, b"sheet"),
+        (image, b"image"),
+        (audio, b"audio"),
+        (notes, b"# notes"),
+        (archive, b"zip"),
+    ):
+        path.write_bytes(content)
+
+    os.utime(report, (100, 100))
+    os.utime(table, (200, 200))
+    os.utime(notes, (300, 300))
+    os.utime(archive, (400, 400))
+
+    rows = [
+        {
+            "thread_id": "thread-one",
+            "display_name": "战场环境综合研判",
+            "metadata": {"agent_id": "analysis-agent"},
+        },
+        {
+            "thread_id": "thread-two",
+            "display_name": "生成发展报告",
+            "metadata": {},
+        },
+    ]
+    store = _FakeThreadStore(rows)
+
+    class _Paths:
+        def sandbox_outputs_dir(self, thread_id: str, *, user_id: str) -> Path:
+            assert user_id == "owner"
+            return tmp_path / thread_id / "outputs"
+
+    async def _resolve_owner(_thread_id: str) -> str:
+        return "owner"
+
+    monkeypatch.setattr(artifacts, "get_paths", lambda: _Paths())
+    monkeypatch.setattr(artifacts, "aresolve_path_user_id", _resolve_owner)
+
+    items = await artifacts._scan_artifact_library(store)
+
+    assert [item["name"] for item in items] == ["证据.zip", "结论.md", "数据汇总.xlsx", "研判报告.docx"]
+    assert [item["file_kind"] for item in items] == ["archive", "text_data", "spreadsheet", "document"]
+    assert items[-1]["thread_title"] == "战场环境综合研判"
+    assert items[-1]["agent_id"] == "analysis-agent"
+    assert all(item["name"] not in {"配图.png", "录音.mp3"} for item in items)
+    assert store.calls[0]["exclude_system"] is True
+
+
+@pytest.mark.asyncio
+async def test_artifact_library_endpoint_filters_searches_and_pages(tmp_path: Path, monkeypatch):
+    output_dir = tmp_path / "thread-one" / "outputs"
+    output_dir.mkdir(parents=True)
+    (output_dir / "任务报告.md").write_text("report", encoding="utf-8")
+    (output_dir / "任务数据.csv").write_text("a,b", encoding="utf-8")
+
+    store = _FakeThreadStore(
+        [
+            {
+                "thread_id": "thread-one",
+                "display_name": "任务研判对话",
+                "metadata": {},
+            }
+        ]
+    )
+
+    class _Paths:
+        def sandbox_outputs_dir(self, thread_id: str, *, user_id: str) -> Path:
+            return tmp_path / thread_id / "outputs"
+
+    async def _resolve_owner(_thread_id: str) -> str:
+        return "owner"
+
+    monkeypatch.setattr(artifacts, "get_paths", lambda: _Paths())
+    monkeypatch.setattr(artifacts, "aresolve_path_user_id", _resolve_owner)
+    monkeypatch.setattr(artifacts, "get_effective_user_id", lambda: "owner")
+    monkeypatch.setattr(deps, "get_thread_store", lambda _request: store)
+    artifacts._artifact_library_cache.clear()
+
+    raw_endpoint = artifacts.list_artifact_library.__wrapped__
+    response = await raw_endpoint(
+        request=SimpleNamespace(),
+        query="研判对话",
+        file_kind="text_data",
+        offset=0,
+        limit=1,
+        refresh=True,
+    )
+
+    assert response.total == 1
+    assert response.offset == 0
+    assert response.limit == 1
+    assert [item.name for item in response.items] == ["任务报告.md"]
+
+    spreadsheet_response = await raw_endpoint(
+        request=SimpleNamespace(),
+        query="任务数据",
+        file_kind="spreadsheet",
+        offset=0,
+        limit=50,
+        refresh=False,
+    )
+    assert spreadsheet_response.total == 1
+    assert spreadsheet_response.items[0].name == "任务数据.csv"

offline-backend-20260512/backend/tests/test_public_llmwiki_embed.py

@@ -68,7 +68,7 @@ def _published_row() -> dict[str, object]:
 
 
 @pytest.mark.asyncio
-async def test_public_knowledge_base_list_exposes_only_published_wiki_identity() -> None:
+async def test_public_knowledge_base_list_exposes_live_safe_metadata_and_generated_index(monkeypatch) -> None:
     store = PublicListStore(
         [
             {
@@ -83,21 +83,116 @@ async def test_public_knowledge_base_list_exposes_only_published_wiki_identity()
                 "owner_user_id": "system",
                 "name": "对话沉淀",
             },
+            {
+                **_published_row(),
+                "id": "stale-kb",
+                "weknora_id": "remote-missing",
+            },
         ]
     )
 
+    class Client:
+        async def list_knowledge_bases(self):
+            return [
+                {
+                    "id": "remote-one",
+                    "name": "远端公开知识库",
+                    "description": "远端公开说明",
+                    "type": "wiki",
+                    "knowledge_count": 12,
+                    "chunk_count": 34,
+                    "processing_count": 1,
+                    "share_count": 3,
+                    "created_at": "2026-07-01T08:00:00+08:00",
+                    "updated_at": "2026-08-01T08:00:00+08:00",
+                    "tenant_id": 10001,
+                    "creator_id": "secret-creator",
+                    "storage_config": {"secret_key": "must-not-leak"},
+                }
+            ]
+
+        async def get_knowledge_base(self, knowledge_base_id: str):
+            assert knowledge_base_id == "remote-one"
+            return {
+                "id": knowledge_base_id,
+                "vector_store_name": "System default",
+                "vector_store_source": "env",
+                "vector_store_engine_type": "elasticsearch",
+                "vector_store_status": "available",
+                "wiki_config": {"enabled": True},
+            }
+
+        async def list_wiki_pages(self, knowledge_base_id: str, **kwargs):
+            assert knowledge_base_id == "remote-one"
+            assert kwargs == {"page": 1, "page_size": 500, "query": ""}
+            return {
+                "pages": [
+                    {
+                        "id": "page-one",
+                        "slug": "concept/start",
+                        "title": "开始使用",
+                        "page_type": "concept",
+                        "status": "completed",
+                        "created_at": "2026-07-02T08:00:00+08:00",
+                        "updated_at": "2026-08-02T08:00:00+08:00",
+                    },
+                    {
+                        "id": "page-two",
+                        "slug": "summary/overview",
+                        "title": "知识概览",
+                        "page_type": "summary",
+                        "status": "completed",
+                        "created_at": "2026-07-03T08:00:00+08:00",
+                        "updated_at": "2026-08-03T08:00:00+08:00",
+                    },
+                ],
+                "total": 2,
+                "page": 1,
+                "page_size": 500,
+                "total_pages": 1,
+            }
+
+    monkeypatch.setattr(llmwiki, "_client_or_503", lambda: Client())
+
     payload = await llmwiki.list_public_knowledge_bases(_request(store))
 
     assert store.calls == [("__public_wiki__", "public", False)]
-    assert payload == {
-        "knowledge_bases": [
-            {
-                "id": "1063edee-332a-43b4-bbd9-73bf0a708b94",
-                "name": "公开知识库",
-            }
-        ],
-        "total": 1,
+    assert payload["total"] == 1
+    item = payload["knowledge_bases"][0]
+    assert item["id"] == "1063edee-332a-43b4-bbd9-73bf0a708b94"
+    assert item["name"] == "远端公开知识库"
+    assert item["description"] == "远端公开说明"
+    assert item["type"] == "wiki"
+    assert item["knowledge_count"] == 12
+    assert item["chunk_count"] == 34
+    assert item["processing_count"] == 1
+    assert item["share_count"] == 3
+    assert item["is_processing"] is True
+    assert item["remote_created_at"] == "2026-07-01T08:00:00+08:00"
+    assert item["remote_updated_at"] == "2026-08-01T08:00:00+08:00"
+    assert item["vector_store_engine_type"] == "elasticsearch"
+    assert item["wiki"] == {
+        "status": "ready",
+        "page_count": 2,
+        "loaded_page_count": 2,
+        "stats_complete": True,
+        "page_type_counts": {"concept": 1, "summary": 1},
+        "page_status_counts": {"completed": 2},
+        "index_mode": "generated",
+        "has_explicit_index": False,
+        "index_slug": "__index__",
+        "default_slug": "__index__",
+        "created_at": "2026-07-02T08:00:00+08:00",
+        "updated_at": "2026-08-03T08:00:00+08:00",
+        "embed_route": "/embed/knowledge/1063edee-332a-43b4-bbd9-73bf0a708b94?wikiSlug=__index__",
     }
+    assert item["local_index"]["state"] == "not_synced"
+    assert item["local_index"]["fully_vectorized"] is False
+    assert item["owner_user_id"] is None
+    assert "weknora_id" not in item
+    assert "tenant_id" not in item
+    assert "creator_id" not in item
+    assert "storage_config" not in item
 
 
 @pytest.mark.asyncio
@@ -178,6 +273,35 @@ async def test_public_wiki_page_list_is_normalized(monkeypatch) -> None:
     assert payload["pages"][0]["aliases"] == []
 
 
+def test_public_wiki_summary_prefers_an_explicit_index_page() -> None:
+    payload = llmwiki._public_wiki_summary(
+        "mapping-one",
+        [
+            {
+                "slug": "guide/start",
+                "title": "开始使用",
+                "page_type": "article",
+                "status": "completed",
+            },
+            {
+                "slug": "home",
+                "title": "知识首页",
+                "page_type": "index",
+                "status": "completed",
+            },
+        ],
+        total=2,
+        complete=True,
+    )
+
+    assert payload["status"] == "ready"
+    assert payload["index_mode"] == "explicit"
+    assert payload["has_explicit_index"] is True
+    assert payload["index_slug"] == "home"
+    assert payload["default_slug"] == "home"
+    assert payload["embed_route"] == "/embed/knowledge/mapping-one?wikiSlug=home"
+
+
 @pytest.mark.asyncio
 async def test_public_wiki_file_is_authorized_and_streamed(monkeypatch) -> None:
     store = PublicStore(_published_row())