Commit Graph

710 Commits (main)

Author SHA1 Message Date
lyingbug a8eb7e28fa
fix(automation): enforce execution identity, cancellation, and scheduling limits (#4506)
* fix(automation): preserve execution identity and scheduler guarantees

* fix(ci): satisfy lint checks for automation regressions
2026-10-20 13:00:12 +08:00
nullptr-error 8dab7550f5
feat(sandbox): 双模式沙箱分配的基础(门控关闭,线上行为不变) (#4507)
* refactor(sandbox): move workspace layout and path rules into a leaf package

internal/sandbox/workspace owns the remote layout (root, input, output,
GitDir, skills root) and the lexical path rules. The sandbox root package
re-exports them through aliases, so callers keep their imports. No
behaviour change.

* feat(sandbox): add the allocation domain model

Pure types for placing a session: session/user mode, the user-mode
admission whitelist (web account or API external user only, with a fixed
fallback reason otherwise), the shared-machine identity and its digest,
the empty-retention bounds and the deployment gate.

* feat(sandbox): add machine and session admission leases

Registry of in-flight operations per shared machine and per session, with
Redis and in-memory implementations. Leases renew only while still live,
report loss at a local deadline measured from the request send time, and
Bind cancels a context when any held lease or barrier is lost. Holding or
renewing only ever extends the scope key's lifetime.

* feat(sandbox): add allocation_mode and empty retention to sandbox configs

Configs gain allocation_mode (session|user, default session) and
user_sandbox_empty_retention_days (0 = default 7, up to 90). Updates that
omit them keep the stored values. Switching a config to user requires the
deployment gate (WEKNORA_SANDBOX_USER_ALLOCATION plus Redis), exposed as
the settings.sandbox.user_allocation capability.

The gate stays closed in this build regardless of the env var: the session
manager does not place sessions by their assignment yet, so a user-mode
placement would be recorded for a session that still runs on its own
machine.

* feat(session): record the server-verified owner auth source

sessions.owner_auth_source records how user_id was authenticated when the
session was created or its owner first bound: web, API direct header or
signed token, API key, MCP endpoint, embed, IM, internal. The auth
middleware puts the source on the context, CreateSession stamps it and
ignores client input, owner binding (MCP ask, embed backfill) records the
entry's own source, and a fork inherits it. Rows from before the column
stay empty, which means unknown.

Per-user sandbox allocation reads it to tell a verified person from a
shared entry; the owner's kind follows from the source and user_id, so it
is not stored separately.

* feat(sandbox): persist session placements and shared-machine registry

session_sandbox_assignments records each session's placement once: its
config pair, the shared machine digest ('' for a machine of its own),
when it was released and when its directory cleanup is due. Mode and
layout follow from the digest. sandbox_user_machines holds the identity
behind each digest and the empty-retention clock.

ClaimPlacement writes the assignment and projects the session pin in one
transaction that locks the session row, then the machine row, before
inserting. RetireSessions releases assignments under the same locks and
writes a tombstone for a session that was never placed, so no placement
lands after a deletion began. The repository is also exercised on the
schema the shipped SQLite migrations create.

* feat(sandbox): place sessions through the allocation service

sandboxalloc decides a session's placement on first remote use and records
it with the pin: sessions that already used a per-session sandbox stay on
it, maintenance sessions are always dedicated, and only an eligible owner
on a user-mode config with the gate open shares a machine. A placed
session resolves straight from its assignment; a released one is refused.
Skill maintenance sessions record a maintenance placement and are removed
if that fails.

* fix(sandbox): retry placement when the sandbox config cannot be read

A failed config load was stored as session mode, so the session could never join a shared machine after the error cleared.

* fix(sandbox): satisfy golangci-lint on the allocation foundations

Wrap lines past 120 characters, comment exported names, and clear the two staticcheck findings so the pull request lint job can pass.

---------

Co-authored-by: hailongzhao <hailongzhao@tencent.com>
2026-10-20 12:28:30 +08:00
lyingbug eecfe4b3c8
feat(automation): scheduled agent runs with in-app notifications (#4492)
* feat(automation): scheduled agent runs with in-app notifications

Automations let a user have an agent run a task on a schedule as them.
The scheduler is database driven: every instance claims due rows with a
version CAS, so Lite and multi-instance deployments share one path.
Runs execute in deferred mode, suspend on approvals and notify the
owner; an answered prompt resumes the run from its snapshot. Standing
grants are promoted from approvals past runs asked for, bound to the
agent configuration and the owner's role.

Adds the automation, run and notification APIs, the automation.*
assistant operations, and regenerates the Swagger document.

* feat(frontend): automations page and notification bell

Adds an Automations page to the sidebar: a list with schedule, agent,
next run and last status; an editor with schedule presets, a custom
cron field, timezone and a live preview of the next runs; and a detail
drawer with run history, links into each run's conversation and the
standing approvals the owner can grant from past approval requests.

The top-right bell now merges pending invitations with in-app
notifications and polls for new ones.

* fix(automation): let the assistant schedule tasks and match the settings UI

The assistant never searched for automation operations, one-time times failed unless they were RFC3339, and the create, detail, and notification surfaces did not match the rest of the app.
2026-10-20 00:11:36 +08:00
lyingbug ba253bcbac
feat(agent): unified agent permission model with grants and run modes (#4490)
* refactor(agent): route approval decisions through a single policy

Operations, assistant write tools, MCP tool approval, delegated-child
approval and the child tool guard each compared access levels on their
own. They now ask policy.Decide, which maps an action's risk and the
run's rules to allow or ask. Behaviour is unchanged: MCP tools still
follow only the workspace per-tool flag and delegation, and a write that
reaches a read-only run still asks.

Add design docs for the agent permission model and scheduled automation.

* refactor(agent): replace interactive flags with a run mode

Runs are attended, deferred or unattended. Unattended runs refuse every
approval, including MCP approvals that used to wait out the timeout;
deferred runs suspend at once.

* feat(agent): apply access levels to MCP tools and every agent

MCP tools get a risk: the workspace's explicit setting, else the service's
readOnlyHint, never lower than a risk the service declared before. The
run's access level now decides MCP calls as it does operations; read-only
runs neither see nor run tools that change data. Ordinary agents get a
configurable ceiling that defaults to auto_write, and delegated children
take the lower of their parent's level and their own ceiling instead of
asking for every MCP call.

* feat(agent): persist approval grants with session/automation scopes

Remembered "allow for this conversation" approvals move from prompt history
into an agent_grants table keyed by root session (or automation binding), so
delegated children share the root grant and unattended runs honour it.
MCP write tools now offer the grant, keyed by service, tool and input schema.
Users can list and revoke their grants from the chat header.

* feat(agent): acting-via identity, decision logs and grant lapse reasons

Principal gains ActingVia so work done on a user's behalf (automations) is
told apart in audit details and policy decision logs. Every non-read policy
decision is logged with its action, risk, level, mode and verdict.
Automation grants bind to a hash of the agent setup, owner role and access
ceiling; a grant that no longer applies asks again and the card says why.

* docs(agent): record permission model implementation status

* chore(handler): check c.Error in agent grant handlers

* chore(handler): check c.Error in MCP tool policy handler

* fix(agent): raise MCP observed risk atomically and scope worker access

- Compare observed_risk inside the upsert so a stale lower observation
  cannot overwrite a higher one recorded concurrently.
- Delegated workers use the same conditional level as attach: the
  assistant ceiling applies only under an assistant parent.
- Mock @/api/assistant in useModalShell tests; gofmt internal/mcp/client.go.

* docs(agent): allow destructive pre-grants only in automation scope

Decide against a level that skips every prompt: Always stays
unwaivable and destructive actions may only be pre-approved per resource
when an automation is set up. Note that automation runs must look up
both automation and session grants.

* feat(agent): standing grant scope and risk-bounded grants

Grants record the risk they were approved for and never cover a riskier
call. Destructive calls are now named for matching, so an automation's
standing grant can cover one, but the card still offers to remember only
writes. The gate looks up the run's own scope and then its standing
scope; approvals during a run are written only to the run's scope.
2026-10-19 20:52:15 +08:00
lyingbug 966d841d19
feat(assistant): add durable multi-agent collaboration and live task views (#4438)
* feat(assistant): add persistent multi-agent collaboration and live task views

* fix(ci): wrap runtime notice message append

* fix(agent): preserve collaboration results and prevent duplicate execution

* fix(agent): fence panic followups and preserve approval timelines

* test(frontend): stub connector labels in assistant refresh fixture

* fix(migrations): avoid collaboration migration collision with main

* fix(assistant): preserve child continuations and artifact access

* fix(frontend): align artifact and icon checks with updated UI

* fix: enforce child approvals and finalize interrupted collaboration runs

* feat: support inherited temporary agent workers

* docs: clarify inherited worker scope and lifecycle

* fix: harden assistant retrieval and final delivery

* fix(agent): keep closed child prompts continuable and renew active deadlines

Expired child approvals were looked up by the child session, so the parent group stayed open and the run became needs_attention. Active collaborations also expired one hour after creation.

* fix(lint): clear golangci-lint findings on the collaboration PR

The previous CI run failed on line length, gofumpt, unchecked closes, and an unused dial parameter.

* fix(ci): use the public npm registry for frontend dependencies

* fix(agent): address collaboration review findings

- keep ordinary fallbacks when a turn delegated no work
- throttle lease/deadline heartbeats and drop duplicated child answers
- force-close or compact aggregates that exceed the size limit
- clamp completion budget to the model output limit
- honor prompt_suspend for non-collaboration tools
- clean up collaborations and child sessions on session delete, prune old rows
- bound concurrent child executions globally
- retry prompt classification and tolerate transient heartbeat errors
- scope prompt history to listed messages, reject nil tenants
2026-10-19 12:14:15 +08:00
linus-liu-web a7d915a70a
feat(resource): 相同图片字节复用一个 handle,不再重复写字节与重复注册 (#4419) 2026-10-17 23:16:40 +08:00
Xinliang Zheng 164dafca85
feat(graph): add standalone JSON graph search API (#4395) 2026-10-17 23:00:55 +08:00
lyingbug 8776a2d92d
feat(assistant): let the assistant find skills and MCP services on the web (#4429)
* feat(assistant): let the assistant find skills and MCP services on the web

Enable web_search/web_fetch for the WeKnora assistant so it can locate a
skill source or MCP endpoint from a description, with prompt guidance on
sources, injected page text, console-only credentials and a missing search
provider. mcp_service.create now always asks the user.

Agent trials no longer stall on prompts nobody can answer: tool approvals
are refused at once and OAuth falls back to the non-interactive notice.

* fix(ci): wrap the unattended approval test constructor

golangci-lint rejects the NewGate call at 127 characters.
2026-10-17 15:25:08 +08:00
lyingbug 0d7516ea0a
feat(assistant): expand management operations and polish interaction UI (#4427)
* fix(assistant): polish interaction cards and stabilize composer context

Localize operation labels across all six UI languages and expose answer copying in the assistant panel. Refine question and page-link cards with a single open/return action, and keep the current-page context row at a stable height.

* feat(assistant): add common data source, chunk and knowledge base operations

* feat(assistant): support parsing configuration and batch FAQ updates

* feat(assistant): add MCP tool directory query and sync operations

* feat(assistant): manage web search providers and data source sync settings

* feat(assistant): add model configuration copy operation

* feat(assistant): add workspace profile members and system information operations

* feat(assistant): add system administrator queries and runtime diagnostics

* fix(frontend): polish assistant and resource card UI

Refine assistant composer layout, context controls and reply actions, including narrow-window alignment. Unify knowledge base, agent and shared-space cards with consistent icons, hover feedback, favorites, menus and compact localized statistics.

* fix(assistant): preserve concurrent settings and exclude upstream error secrets
2026-10-17 14:35:58 +08:00
lyingbug 7115cb2a61
feat(assistant): trial runs, session grants, page links, and MCP/model/skill management (#4375)
* feat(assistant): let the assistant check MCP services and retrieval

Add a probe risk for operations that change nothing but do real work
(connecting to a service, running a retrieval). Probes use POST, run
without approval at every access level, and are capped per run.

Open three operations on top of it:
- mcp_service.test: connect with the saved settings and report why not
- mcp_service.list_tools: list what a service offers before binding it
- knowledge_base.search: raw hybrid recall with thresholds, to explain
  why a document is or is not found

x-agent can now name one method when a handler documents several, and
catalog parameters accept numbers.

* feat(assistant): let the assistant try an agent it set up

Add try_agent: the assistant asks an agent one question the way a user
would and gets back its answer, the tools it called and the documents it
cited, so it can check a setup before calling it done.

- The trial runs as the signed-in user in a new agent_trial session kind,
  which ordinary listings already exclude and the API cannot create, and
  the session is removed afterwards. Runs are non-interactive, so tools
  that would ask the user are refused. Long-term memory is untouched: it
  is written only by the HTTP handler.
- The assistant itself, the Wiki fixer and the skill installer cannot be
  tried, nor any agent granted the operation catalog.
- A trial spends model calls and runs the agent's own tools, so it is a
  write for the access level: confirm asks, auto_write runs, read_only
  does not offer it. The confirmation card names the agent and question;
  ConfirmedTool now lets a wrapped tool supply that text.
- At most 3 trials per turn, 4 minutes each; the tool timeout is raised
  above that. Answers, tool errors and references are capped.
- The runner is bound to the session service after the container builds
  it, since the session service depends on the agent service.

* feat(assistant): let the user allow similar changes for the conversation

The confirmation card for a recoverable change now offers "allow this for
the rest of the conversation". Once chosen, later calls of the same kind
in the same session run without asking, so a batch tidy-up no longer
needs a click per item.

- What a grant covers is derived on the server, never taken from the
  model: the same catalog operation on the same knowledge bases (a
  document's knowledge base comes from the target read) or the same
  agent; the same Wiki tool on the run's Wiki knowledge bases; trials of
  the same agent. Deletions and operations without such a scope, such as
  creating a knowledge base, always ask.
- No new storage: the gate looks for an earlier approved prompt in the
  session with the same grant and allow_in_session set, for the same
  tenant and user. Grants therefore survive suspend/resume and replicas,
  and end with the conversation. A failed lookup asks again.
- The response validator accepts allow_in_session only on approval and
  only when the prompt offered a grant, so a client cannot widen it.
- The card shows the scope in the user's language. The model is told
  such calls may run without a card and that deletions always ask.
- describeTarget now returns its target read so the grant reuses it
  instead of reading again.

* feat(assistant): tell the assistant about settings pages and chat answers

Pages now report which page is open and, for the agent editor, which
section. The model and MCP settings name the model or service open for
editing, and chat answers gain an "ask the assistant" button that
attaches that answer. The backend resolves these for the assistant only:
models and MCP services in the caller's tenant without endpoints or
credentials, and messages through the caller's session read access, with
the question, an answer excerpt and cited sources quoted as data.

* feat(assistant): link the user to pages and filled-in forms

The assistant gains open_page, which shows a button to a knowledge base
or document, a Wiki page, an agent's editor at a tab, a model or MCP
service to edit, or a new model or MCP service with its form filled in.
Targets are read through the operation catalog as the user, so a link
never points at something they cannot open. Drafts take no credentials
and only http(s) endpoints without user info; the card shows the
endpoint, and the user enters keys and saves the form themselves.

* feat(assistant): manage MCP services, model names and skills

* fix(assistant): prefill model drafts on initial navigation

* feat(assistant): document tags, model catalog reads, clearer cards, live page refresh

- Operations project lists of objects; knowledge.get/list report tags.
- model.get shows provider, endpoint and capabilities; new
  model.list_providers and model.get_provider read the model catalog,
  including vendors an admin added.
- open_page and ask_user guidance: offer page buttons whenever the next
  step is the user's, word choices for the user.
- Chat cards: redesigned open-page card pinned under the final answer,
  ask-user inputs in the dock style, approval card with aligned fields,
  translated labels and a question built from the operation summary.
- Open pages reload after assistant changes: deletes and moves wait for
  the background task, Wiki edits now report their changes, KB list
  counts, tags and chat model selectors refresh.
- The assistant panel stays open across reloads within the tab.

* fix(ci): format assistant operation code and wrap long lines

* feat(assistant): let the assistant see, fill and open the page beside it

- read_page / fill_page tools answered by the browser over a page_action prompt,
  with refs, redaction, in-place text edits, undo card and a page-access switch
- composer shows the open page, dialog and selection; a selection bubble and an
  editor toolbar entry ask the assistant about the selected text
- edits scroll into view with a highlight and an inline undo bar
- open_page reaches any settings section, toolbox tool and list page, and can
  navigate there itself to read it, keeping the button when edits are unsaved

* feat(assistant): let the assistant open the tab it needs

A visible tab can be selected and re-read, and system settings can open straight to user management, so the next step is no longer blocked on the wrong tab.

* feat(assistant): expose read-only conversation operations

Add session.list, message.list, message.search and
message.chat_history_stats to the operation catalog, restrict the
sessions and messages areas to non-mutating operations, and search
messages with the raw query instead of the log-sanitized one.

* style(chat): restyle open-page and page-edit cards

* fix(ci): comply with assistant card styles and Go line limits
2026-10-16 23:36:01 +08:00
Greenplumwine 973d9d051f
feat(datasource): deep-adapt Feishu/Lark sync with folder mirror, block parsing and embedded objects (#3497)
* ci(build): add manual docker build workflow

Co-Authored-By: Claude <noreply@anthropic.com>

* feat(datasource): deep-adapt Feishu/Lark sync with folder mirror, block parsing and embedded objects

- folder mirror: wiki/drive connectors emit path-qualified FileName so
  knowledge lands in KB folders mirroring the Feishu tree; shortcut
  subtrees deduped; sanitized directory names; legacy datasources
  converge via one-shot resync marker (incremental sync upgrades to
  full once, then clears)
- parse mode: replace FEISHU_DOCX_PARSE_MODE env with per-datasource
  settings.parse_mode (blocks default, export escape hatch); frontend
  selector + 5 locales; zero-migration read-time fallback
- block-level GFM: extend docx block coverage (callout/grid/iframe/
  board/mindnote/agenda/synced blocks), rich text elements (links,
  inline code, KaTeX equations, mentions), real ordered-list numbering,
  code language mapping, merged-cell fill, truncation note
- embedded objects: images (gated off multimodal toggle) and boards
  (download_as_image, 403-degraded) stored as knowledge with
  weknora-img://<n> markers resolved in the document-processing
  pipeline via knowledge FilePath; document attachments whitelisted
  (pdf/md/xls(x)/csv/doc(x)/ppt(x)) as standalone knowledge in the
  parent folder with bidirectional parent/child metadata and filename
  list in the body; videos skipped with placeholder + source link
- golden baseline explicitly re-recorded; docs (.env.example,
  docker-compose, docs/*, website-docs) synced

* fix(datasource): address Feishu deep-adaptation smoke-test findings

- folder mapping: drop the space/root-folder name prefix — the selected
  sync root maps to the knowledge-base root, subdirectories stay relative
- docx blocks: render view(33) blocks silently as containers (children
  only), removing the stray '[飞书块: 视图]' placeholder before file lists
- sheet inline tables: fetch merge ranges via sheets/v3 sheets/query and
  fill the top-left value across the merged region (closed-interval
  interpretation, flagged for real-environment verification)
- knowledge list: hide embedded image/board rows (metadata
  embedded_image=true) from the document list and folder-tree counts;
  rows remain for VLM indexing and marker resolution; search untouched
- frontend: hydrate protected images (resource:// etc.) in the document
  preview so embedded Feishu images/boards actually render

* fix(datasource): resolve embedded image markers at ingest time

The weknora-img://<n> markers were only resolved during document
processing, so chunks carried real resource:// URLs while the stored
markdown file and source_content kept raw markers — file preview
showed blank images. Resolve them in ingestItem before
CreateKnowledgeFromFile (children are emitted first, so image rows
with FilePath already exist); markers with missing children are kept
for the ProcessDocument fallback to degrade. Stored file,
source_content and chunks now agree. Existing datasources converge on
the next full re-sync (update=delete+recreate).

* feat(datasource): resolve Feishu mention users to real names at ingest

docx mention_user elements only carry an open_id, so they rendered as
a generic '@成员'. Resolve the display name via
GET /contact/v3/users/:id (user_id_type=open_id) while generating the
markdown during FetchDocxWithBlocks — before storage, per user
decision; the frontend performs no lookups. A per-sync closure caches
positive and negative lookups; permission errors and empty names
silently degrade to '@成员' (debug log only, ingest never blocked).
Requires one of the contact read scopes (e.g.
contact:user.base:readonly) on the Feishu app.

* refactor(datasource): polish round — defect fixes and structural cleanup

- attachment placeholder patching now targets the unique
  '- weknora-att://<token>' marker line rendered per attachment instead of
  matching by display name, so an identical earlier bullet is never
  replaced by mistake (regression-tested)
- markdown link targets with spaces or parens are percent-escaped
  (iframe / mention_doc / inline links), keeping them parseable
- drive folder-tree walking consolidated into core.WalkDriveTree,
  deleting the 174-line copy in drive/paths.go
- parse mode threading made explicit: DocxFetchInput.ParseMode replaces
  the private client field (connector entry resolves via
  ParseFeishuConfig and passes it through ops)
- maxFeishuDownloadBytes restored to a const; the oversize-degrade test
  now declares an over-cap Content-Length and downloadRawBytes rejects
  early on the header

* refactor(datasource): align Feishu image handling with the platform image pipeline

Replace the image-as-separate-knowledge-row channel with the platform's
orthodox flow: the connector downloads image/board bytes at fetch time and
embeds them in the parent markdown as base64 data URIs; the document
processing image resolver (image_resolver.ResolveAndStore, the same stage
used for parsed PDFs/Word) stores them on the tenant file service, swaps in
persistent provider:// URLs and produces OCR/caption child chunks when a VLM
is configured.

Removed with the marker pipeline:
- weknora-img:// ingest resolution in ingestItem
- the ProcessDocument Step-1.9 marker hook
- the document-list/folder-tree filter for embedded_image rows
- the feishu-specific runtime parse_mode backfill; legacy data sources are
  now stamped by migration 000096 (sqlite 000017) which writes the one-shot
  resync_required marker — the generic full-upgrade + clear consumption stays

Attachment (PDF/Word/md) sub-items are unchanged.

* style: gofmt test files

* fix(datasource): resolve Drive sub-selection kind from the root listing

A single-file Drive selection (rootFolderToken:fileToken) was probed with
folder APIs before falling back: folder meta returned 91202 and the folder
list 1061002 on every sync, logging API errors for a working configuration.

listDriveFilesForResource now walks the root subtree first and reads the
selected node's kind off the listing: a file filters straight out of the
already-fetched results (zero error calls), a sub-folder still gets its
dedicated walk, and only a token missing from the listing (partial listing /
permission edge) keeps the legacy probe + fallback.

Also pins sub-folder selection semantics with a dedicated test.

* fix(datasource): address review findings across the Feishu deep-adaptation branch

Four parallel reviewer slices audited the branch; this applies the accepted
findings.

Content fidelity (blocks rendering):
- todo blocks parse style.done — completed items render "- [x]" instead of
  always unchecked
- code fences size past the longest backtick run in the content, so a line
  of ``` inside code can no longer close the fence early
- ordered-list "auto" numbering only resets on a sibling of the last item;
  child blocks rendered flat between items no longer renumber the run

Image pipeline:
- inline base64 images are capped at the document-processing resolver's
  10 MB budget; larger images degrade to a placeholder instead of leaving a
  permanent raw-base64 blob in stored markdown
- image markers carry a per-document nonce so user-typed look-alike text
  can never be patched in place of a real marker
- attachment marker patching anchors on the token's line end, closing the
  token-prefix collision
- SupportedImageExt recovers image/svg+xml (board exports), which the
  platform resolver already stores natively

Sync correctness:
- clearResyncMarker re-reads the row before writing: a long upgraded sync
  no longer risks reverting concurrent user edits with a stale full-row
  update
- wiki EditTime keys on max(obj_edit_time, node_edit_time): pure
  moves/renames re-file the document on the next incremental sync
- file-typed wiki/drive base names are sanitised; a "/" in a title can no
  longer fake an extra folder level

Docs/tests/comments:
- datasources docs rewritten to the v3 image contract (no image rows,
  no image_map, no parse_mode backfill claim)
- DataSourceEditorDialog drops settings.parse_mode when backtracking to a
  non-Feishu connector type
- stale comments corrected (types.go parse-mode transport, repository
  folder-counts, wiki List partial-error branch, test headers); regression
  tests added for todo done, fence sizing, ordered numbering, svg sniff

* fix(datasource): repair markdown conversion defects found in live smoke test

Live re-sync of the all-block-type Feishu test document surfaced three
converter defects:

- Code blocks: the closing fence reused the opening fence string, so a
  tagged block ended with ```java — CommonMark requires a bare closing
  fence, and every renderer left the block unterminated. The closing
  fence is now backticks only.
- Feishu heading7-9 emitted 7-9 hash marks; GFM stops at H6, so those
  lines rendered as literal '#######' text. They now degrade to bold.
- The docx API percent-encodes link targets (http%3A%2F%2F...); escapeURL
  now reverses that encoding before markdown-escaping, so stored links
  point at real URLs instead of percent-encoded strings.

Verified against the live document via a throwaway API probe: the
'i|g|k' table cells questioned during review are the document's own
content (the sheets copy says i|j|k; the two docx tables genuinely
contain 'g'), not a converter bug.

* fix(frontend): render GFM task lists in document preview

Task list items (- [x] / - [ ]) rendered as plain bullets: marked emits
<input type="checkbox"> for them, but the preview DOMPurify config's
ALLOWED_TAGS allowlist has no 'input', so the element was stripped whole.
The preview renderer now overrides checkbox to emit a styled
task-checkbox span (class/attr both allowlisted) and the preview styles
draw the box with a check mark, hiding the list bullet via :has().

Same gap exists in the chat markdown surface; left untouched there.

* fix(datasource): crop blank canvas borders from whiteboard exports

The board download_as_image API renders the WHOLE board canvas, not the
content bounding box — editors routinely leave the canvas far larger than
the drawing, so board/mindnote/diagram exports surfaced with large blank
areas below the content. The API has no crop parameter, so the connector
now trims near-uniform blank borders after download (bottom-right pixel
seeds the background color, 14/255 per-channel tolerance absorbs JPEG
noise, 8px margin kept). PNG and JPEG are re-encoded; anything
undecodable, fully blank, or already tight passes through unchanged, and
oversized images (>64MP) skip the scan to bound memory.

* fix(datasource): require corner majority before trimming board exports

The blank-border trim seeded the background color from the bottom-right
pixel alone. A large solid shape reaching that corner would seed the
background with its own fill color, and fully-filled rows could then be
trimmed as blank — cutting real content. The background color is now the
majority of the four corner pixels (>=3 must agree); on a split vote
content reaches the corners and the image passes through untrimmed.
Content visible against the background is untouched either way: a row or
column survives if even one pixel differs from the background by more
than the 14/255 JPEG-noise tolerance.

* fix(datasource): widen board trim margin and lower JPEG quality to 80

The 8px margin read as a tight crop — a 32px white frame now visually
bounds the content region. JPEG re-encode quality drops 90 -> 80 to keep
the re-encoded size closer to the original export.

* feat(datasource): timeline widget Markdown degradation + table column alignment

Two content-fidelity gaps from the live smoke test:

- 文档小组件 (add_ons, block_type 40) carry their full data inline in
  record; there is no server-side render to download. The timeline
  component (component_type_id blk_6358a421bca0001c...) now degrades to
  a Markdown list (- **time · title**:text), honoring contentShow and
  omitting empty segments; unknown component types and malformed records
  keep the previous placeholder note.
- docx text blocks carry style.align (1 left / 2 center / 3 right,
  absent unless the author set it). Native-table rendering now collects
  per-column alignment from the first explicit cell and emits GFM colon
  separators (:---: / ---:). Embedded sheets/bitable keep the plain
  separator — their value APIs expose no alignment.

* feat(datasource): show document titles for 云文档 mention links

The docx renderer emitted mention_doc references as [url](url) because
the payload carries no title. Feishu doc URLs now emit a
[weknora-docname://token](url) placeholder and FetchDocxWithBlocks
resolves the titles in one drive metadata batch call
(POST /drive/v1/metas/batch_query, docx/doc/sheet/bitable/wiki/...),
patching the link text to [文档名](url). Refs whose metadata cannot be
fetched — no permission (970003), deleted/mismatched (970005), API
absent — degrade to the previous URL-as-text form with a warn log.

This is display-only: referenced docs are not pulled as sub-documents
(deferred design lives in the task PRD).

* docs(datasource): document mention rendering enhancement and drive rename limitation

- Add optional mention scopes (contact:user.base:readonly,
  drive:drive.metadata:readonly, board:whiteboard:node:read) with
  degrade behavior to the Feishu permission tables
- Describe mention_user real-name lookup, mention_doc title backfill
  and timeline widget placeholders in the blocks rendering flow
- Document the Drive known limitation: folder renames / file moves do
  not bump modified_time, so incremental sync keeps the old KB folder
  structure until a full sync; Wiki sources are unaffected
- Clarify the directory-mapping wording in website-docs to distinguish
  Wiki (rename-aware) from Drive (rename-blind) sources

* docs(datasource): add blocks vs export tradeoff summary to website-docs

* docs(datasource): document docx block-type coverage for blocks mode

- Add a full block-type support table (block_type 1-52 grouped) to the
  datasource dev doc: text/heading/list/task/quote structures, code
  fences with language aliases, containers (callout/quote/grid/view),
  native tables with merged-cell splitting (top-left value fill) and
  column alignment, embedded sheet/bitable converted to in-document
  Markdown tables, image/board inline pipeline, attachment whitelist,
  iframe links, timeline add-on degradation and the placeholder set
- Summarize the same coverage in the drive user guide, the wiki
  quick-start doc and website-docs; fix the stale claim that board
  blocks degrade to placeholders (they ride the image pipeline now)

* style: resolve diff-scoped lint findings across feishu packages

- Wrap over-length signatures and calls (lll, max 120)
- Add doc comments to exported Block* payload types (revive)
- Rename unused closure parameters to _ (revive)
- Check w.Write / resp.Body.Close error returns (errcheck)
- Apply De Morgan rewrite in docname token validation (staticcheck)
- gofumpt-format touched files

golangci-lint run --new-from-rev=main ./... now reports 0 issues.

* ci: drop fork-only docker-build workflow

The manual docker-build workflow lives on the fork's main and is not
part of the upstream contribution; it entered the branch through the
fork-main merge.

* fix(migrations): renumber feishu resync migrations to avoid version collision with upstream

Upstream added 000017_dingtalk_stream_only (sqlite) and 000096_dingtalk_stream_only
(versioned) before our merge, colliding with our 000017/000096 feishu resync
migrations. golang-migrate rejects duplicate versions. Renumber to the next free
slots (sqlite 000027, versioned 000108) and update comments + test constant.

* fix(migrations): renumber feishu resync migrations ahead of upstream 000027/000108

Upstream main now also claims sqlite 000027 and versioned 000108
(session_sandbox_config_tenant), colliding with our just-renumbered
resync migrations. Move to 000028 / 000109.

* fix(feishu): address review — deprecated env fallback and upgrade notes

- Keep FEISHU_DOCX_PARSE_MODE as a one-release deprecated fallback when a
  data source has no per-source parse_mode (deprecation warning logged);
  annotate it in .env.example / docker-compose.yml instead of removing.
- Point stale migration-number comments at the final numbers
  (versioned 000110 / sqlite 000030) in datasource_service.go and tests.
- website-docs: add Feishu upgrade notes (env deprecation, default mode
  flip to blocks, one-shot full resync with embedding/VLM recompute);
  refresh 24-feishu-drive.md parse-mode section to the per-source
  setting and new blocks semantics.

* docs(feishu): document necessary and optional connector permissions

Restore the permission matrix lost with the docs/ removal: necessary
scopes (wiki/drive/export/docx) with missing-permission symptoms and
optional enhancement scopes (embedded sheets/bitable, board export,
mention user/doc rendering) with their degrade behavior, in
10-datasource.md's Feishu section; make 24-feishu-drive.md list the
three necessary drive scopes and link to the full table.

* Revert "docs(feishu): document necessary and optional connector permissions"

This reverts commit 3be19cb1b656b3596f2cb6afb2cda8c2a1d12be1.

* docs(feishu): permission matrix by connector, parse mode and necessity

Necessary-scope tables split by connector (feishu/lark wiki vs
feishu_drive/lark_drive) with blocks/export applicability columns and
missing-permission symptoms; optional enhancement scopes with their
degrade behavior; scope names cross-checked against the Feishu open
platform API docs (files/folder list, medias download, export tasks,
metas batch_query). Full matrix lives in 10-datasource.md; the drive
guide mirrors its own table and links back.

* test(feishu): adapt mention_user_test to upstream retryTestServer signature

* feat(feishu): mirror the sync root (wiki space / drive folder) as a top-level KB folder

So a knowledge base linking several feishu sources can tell them apart at a
glance: syncing folder 测试1 yields 测试1/... and syncing wiki space
飞书知识库同步测试 yields 飞书知识库同步测试/.... Root-name resolution
failure falls back to the previous root-less behaviour without failing the
sync; existing sources converge through the one-shot resync marker.

* refactor(feishu): cache wiki space folder prefix per run, refresh stale comments

Review follow-ups for d012e593f: cache the sanitized space prefix per spaceID
for the whole sync run (K selected spaces now cost one ListWikiSpaces paging
pass instead of K), fix the misleading warn wording, and realign the
WalkDriveTree / driveFolderName doc comments with root-mirror semantics.

* docs(feishu): drop stale and duplicated comments from the deep-adaptation

- board.go: trimBoardImage no longer claims a bottom-right seed (the code
  votes the background over the four corners)
- shared.go: rewrite the FetchDocxWithBlocks note (images/boards inline via
  base64 now; removes a stale-marker splice artefact)
- markdown.go: drop an internal task-number leftover
- datasource_service.go: collapse the ProcessSync/processSyncStreaming notes
  duplicated verbatim from resyncRequired's doc comment, keep the call-site
  why (marker cleared only on success)
- datasource_resync_test.go: drop the obsolete 'backfill' wording

* feat(i18n): 添加解析模式及其提示信息到繁體中文語言包

* fix(frontend): use radius token for task checkbox; fix simplified chars in zh-TW parse mode

CI guard fixes: the PR-added task-checkbox style pushed the radius-literal
count to 161 (> baseline 160); the new zh-TW keys missed the locale parity
audit (user's dc6fdcff9) — also normalise the simplified 文档 leftovers.

* fix(feishu): preserve sync scope, nested content, and legacy settings

Cover shortcut selection, ancestor path changes, nested containers and table boards with regression tests. Avoid redundant Drive traversal, repair resync migrations for legacy settings, and preserve inherited parse modes. Bound board image dimensions before decoding.

* docs(feishu): clarify shortcut scope and incremental path tracking

* Fix Feishu sync reconciliation and embedded media regressions

* Bound Feishu table rendering by actual cell data

---------

Co-authored-by: Claude <noreply@anthropic.com>
Co-authored-by: wizardchen <wizardchen@tencent.com>
2026-10-16 16:49:44 +08:00
tayadwx db7046b85e
refactor(image): turn the hard-coded image flow into configurable pipelines (#3746)
* feat(image): configurable image pipelines, rebuilt on current main

Rebuilds the image-pipeline work on top of the current main, carrying
only the pipeline and request-side changes:

- The hard-coded image flow becomes registry-driven pipelines (default /
  smartocr), selectable per knowledge base, with a frontend panel and
  per-action thinking switches addressed as action-level keys
  (caption.thinking / ocr.thinking) declared once and carried into every
  pipeline.
- Request-side support: VLM.PredictWithOptions carries a per-call
  thinking switch through every implementation and decorator, including
  the background concurrency wrapper.
- OCR failure handling follows main's #4132 semantics: invalid output is
  rejected and recorded (ocr_status / ocr_error_code / .ocr subspan)
  without retrying the whole image; caption/observation failures get
  symmetric subspans instead of leaving the image span green.
- Permanent VLM failures skip the asynq retry budget; a run that
  produced nothing fails honestly instead of passing as skipped.
- UI and locale fixes travel with the panel.

Concurrency, rate limiting and circuit breaking are NOT part of this
branch: they stay with the existing system limiter.

* fix(image): address pipeline configuration and execution regressions

* fix(image): align pipeline overrides and validation semantics

* style(image): keep the new pipeline code inside the lll and gofumpt gates

---------

Co-authored-by: tayadwx <tayadwx@users.noreply.github.com>
Co-authored-by: wizardchen <wizardchen@tencent.com>
2026-10-16 16:18:57 +08:00
lyingbug 2fbfb44715
feat(assistant): let the assistant manage knowledge bases (#4353)
* feat(assistant): let the assistant manage knowledge bases

Create, update, and delete knowledge bases, tags, and manual Markdown,
and show those confirmations as labeled rows with model names.

* fix(assistant): satisfy lint on knowledge base operations

Check new error returns, wrap lines past 120 characters, and document the exported tag request types.

* fix(assistant): enforce knowledge base settings permissions and preserve patch fields
2026-10-15 23:54:14 +08:00
lyingbug 4d4acec418
feat(assistant): manage agents and skills, separate history, refresh pages (#4344)
* feat(assistant): let the assistant manage agents

Expose custom agents to the operation catalog (list, get, create, copy,
update, delete) plus read-only model and MCP service lookups, so the
assistant can set up agents end to end.

- Add PATCH /api/v1/agents/:id. PUT replaces the whole config, so a
  one-field change from the assistant would wipe the rest; PATCH only
  overlays the fields it carries and reuses the PUT validation path.
- Add x-agent-detail so long fields such as the system prompt are only
  projected for single-object reads, not lists.
- Add label_param so a create without a target is labelled by its name.
- The approval card lists agent setting changes and resolves model,
  knowledge base and MCP IDs to names.
- Deny sharing, publishing channels, credentials and OAuth routes; model
  and MCP endpoints, headers and secrets are never projected.
- Update the assistant prompt, starters, API docs and Go SDK.

* feat(assistant): list skills and configure agent skill scope

- Expose usable skill names for an agent sandbox without exposing skill
  administration or connection settings to the model.
- Validate PATCH selections against the workspace sandbox usable set so
  invented names and missing sandbox selections cannot be saved.
- Show skill changes in approvals and align the assistant prompt, API
  documentation, and Go SDK with focused refusal and projection tests.

* feat(assistant): isolate conversation history

- Keep assistant conversations out of ordinary session lists, search, artifacts and delete-all cleanup.
- Add private, paginated assistant history with session switching, confirmed deletion and stale-session recovery.
- Backfill session kinds safely on Postgres and SQLite, preserve kinds on forks, and align API docs and SDK coverage.

* feat(assistant): refresh pages after successful data changes

- Carry decoded mutation targets through live tool-result events without
  exposing refresh metadata to the model or persisted chat history.
- Coalesce page and selector refreshes, update document drawers, and warn
  about changed agents while preserving open editor drafts.
- Cover mutation, approval, stream, lifecycle and cache behavior with Go
  and Node tests using the existing frontend runner and component fixtures.

* feat(assistant): add agent page context and resource scope limits

- Attach verified agent context and persist personal scope across turns and approvals.
- Enforce resource boundaries before operation and Wiki approvals, including edited arguments.
- Add scope controls, translations, API and SDK support, and regression coverage.

* refactor(assistant): remove the scope limit

The scope picker confused users and duplicated what RBAC and the access
level already enforce. Keep the agent page context: the agent editor
still publishes agent_id and the backend still verifies it.

* feat(assistant): render operation and product-doc results

Show find/run operations and WeKnora docs as structured cards instead of raw JSON.

* feat(assistant): default new agents to smart-reasoning

Quick-answer cannot call MCP, and agent.update could not change the mode. New agents now start in smart-reasoning, and a patch can switch agent_mode.

* fix: resolve assistant management CI failures
2026-10-15 21:43:20 +08:00
lyingbug 450b6ed1b7
feat(assistant): side panel, access levels and role-aware operations (#4342)
* feat(frontend): add dockable assistant side panel

* feat(assistant): access levels, role-aware operations and assistant settings

* fix(assistant): preserve approval and workspace access safeguards
2026-10-15 18:18:22 +08:00
lyingbug a730a411d4
feat(agent): let the assistant manage FAQ entries (#4327)
* feat(agent): let the assistant manage FAQ entries

New operations: faq.list_entries, faq.get_entry, faq.create_entry,
faq.update_entry, faq.add_similar_questions and faq.delete_entries
(up to 20, each question named on the approval).

The catalog learns what FAQ needs:
- integer path parameters (entry_id) and integer lists (entry_ids);
  approval items resolve integer IDs too;
- x-agent-required, which makes the model send a field the API treats
  as optional. The FAQ update replaces the whole entry, so a left-out
  answer list, tag or answer strategy would be cleared;
- a body on DELETE, which the FAQ batch delete uses.

The FAQ list documents its page of FAQEntry instead of an untyped page,
and the delete body is a named type. FAQ tags are referenced by seq id,
so knowledge_base.list_tags also returns tag_seq_id.

* fix(agent): resolve FAQ operation lint failures

* fix(agent): preserve complete FAQ content during updates
2026-10-15 16:14:21 +08:00
lyingbug aaa1accebd
feat(agent): let the assistant delete and move documents in batches (#4319)
* feat(agent): let the assistant delete and move documents in batches

A batch target can now resolve each item: x-agent "items" names the list
parameter and a read run once per item, and the approval lists every
title under the knowledge base name. The list must be capped at
MaxApprovalItems (20) by x-agent-maximum, which now also bounds string
lists, so nothing approved goes unnamed. One item the user cannot read
blocks the whole batch before any prompt.

New operations: knowledge.batch_delete, knowledge.move (to another
knowledge base, destination named on the card),
knowledge.get_move_progress and knowledge_base.list_move_targets.
The move and move-progress Swagger responses now show their data
envelope.

The approval card keeps line breaks in its description and renders
trailing "- " lines as a scrollable list.

* fix(agent): reconcile batch approvals with workspace operation updates

* fix(agent): keep every batch approval item visible
2026-10-15 15:57:58 +08:00
lyingbug a37a1bc51a
feat(agent): open more workspace operations to the WeKnora assistant (#4318)
* feat(agent): open more workspace operations to the WeKnora assistant

Reads: knowledge_base.list / list_shared / list_tags, knowledge.list_folders,
chunk.list, wiki.get_page / list_folders / list_revisions; knowledge.list
also takes tag_ids, folder_path and folder_recursive.

Recoverable writes: knowledge.update (title, summary), batch_reparse,
move_to_folder, rename_folder, create_manual, create_from_url,
knowledge_base.generate_profile, wiki.create_folder / move_page / revert.

x-agent gains catch_all for routes registered as *slug, which Swagger
cannot express. run_operation decodes knowledge_ids handles. The shared
knowledge base list, the tag page and the URL import body get concrete
Swagger types instead of free-form objects.

* fix(agent): show the assistant the change it is about to make

Manual documents can reach the API length limit, unpaginated lists cover
a tenant, and tag pages report their total. wiki.list_revisions can
return one stored revision before revert, and the confirmation card
names the URL, folder, version and documents.

* fix(agent): align catalog tests with the knowledge API allowlist

Catch-all assertions follow the knowledge routes, and the nested tag
page fixture stays inside an area the catalog accepts. The rune-limit
error also fits the line-length check.
2026-10-15 15:19:23 +08:00
lyingbug 1b24132566
fix(agent): let the WeKnora assistant search without its own rerank model (#4315)
* fix(agent): let the WeKnora assistant search without its own rerank model

Every assistant run with a knowledge base in scope failed with "rerank
model is not configured: please set rerank_model_id on the agent". The
assistant lists search_knowledge, and agents with that tool must name a
rerank model; but the assistant is hidden from agent management, so no
one can set one.

Internal built-in agents now take the workspace's rerank model: its
default active one, else its first active one. With none, search_knowledge
keeps retrieval order, as it already does when reranking fails.
User-configured agents still have to name their rerank model.

UsesFallbackChatModel becomes UsesWorkspaceModelFallback, since it now
covers both models.

* fix(ci): wrap assistant rerank fallback log calls
2026-10-15 14:45:40 +08:00
lyingbug 7c9088d0fe
feat(agent): let the WeKnora assistant read the bundled documentation (#4303)
* feat(agent): let the WeKnora assistant read the bundled documentation

Embeds the website-docs pages (01-07) into the server and adds
search_weknora_docs / read_weknora_doc for the built-in assistant, so it
answers how WeKnora works from the docs of the running version.

- pages split into sections with the site's VitePress anchors; Mermaid
  dropped, screenshots reduced to captions, links rewritten to public URLs
- in-memory BM25 over sections, with Han character pairs and whole
  identifiers such as OLLAMA_BASE_URL or v0.8.3
- granted to the built-in assistant by ID, like the operation tools; the
  docs are public, so no signed-in user is required
- tests check every embedded page parses and every internal link resolves

* fix(agent): read long doc sections to the end and keep docs on a broken page

- read_weknora_doc takes an offset: a section too long for one read is cut
  at a line break and returns next_offset, so leaf sections such as the
  model field reference are readable to the end instead of losing the tail.
- Reads follow the registry's output budget (capped at 8000 runes).
- Percent-encoded anchors, as a browser copies them, and http:// doc URLs
  resolve.
- productdocs.Load leaves out a page it cannot parse and reports it, so one
  broken page no longer withholds the documentation tools.
- App CI runs on changes to the embedded doc pages and the sidebar config,
  where the parse and link tests now apply.
2026-10-15 13:06:58 +08:00
lyingbug 687b178706
feat(agent): add the built-in WeKnora assistant with an operation catalog (#4302)
* feat(agent): add the built-in WeKnora assistant with an operation catalog

Adds builtin-weknora-assistant, which inspects and manages WeKnora with the
signed-in user's permissions:

- find_operations / run_operation over an allow-listed API operation catalog,
  dispatched in-process through the real router as the user
- builtin approval for write/destructive operations, refused when
  non-interactive
- agent attribution in knowledge base activity, including async tasks
- page_context request field, narrowed to the turn's search targets and
  rendered into runtime_context with model handles

* refactor(agent): compile the operation catalog from x-agent Swagger annotations

The catalog was a hand-written copy of what the swag annotations already
describe: routes, parameters, enums and response shapes. It now compiles
from the Swagger document embedded in the binary. Handlers opt in with

  // @x-agent      {"id":"knowledge.delete","risk":"destructive",...}
  // @x-agent-hint "Cannot be undone."
  // @Param        id path string true "知识ID" extensions(x-agent=knowledge_id)

and response types mark the fields the model may see with an
extensions:"x-agent" struct tag. Exposure stays opt-in at every level:
unmarked operations, parameters and fields are not reachable, so a new
field on Knowledge or a new query parameter never reaches the model by
accident. NewCatalog still enforces the safety rules (forbidden areas,
risk versus method, approval targets), and a test fails when the
annotations and docs/swagger.json disagree.

The twelve operations compile to the same routes, parameters and
results as before; document results now share one field set, and the
parse_status filter accepts deleting, which the API supports.

* fix(agent): enforce resume and wiki context scopes
2026-10-15 12:18:30 +08:00
lyingbug 234aa948db
docs(swagger): document the real request and response types (#4304)
* docs(swagger): add shared response envelope types

* docs(swagger): type knowledge base responses

* docs(swagger): add concrete response body types

* docs(swagger): type chunk list responses

* docs(swagger): type knowledge mutation responses

* docs(swagger): type wiki search and autofix responses

* docs(swagger): type FAQ responses

* docs(swagger): type tag responses

* docs(swagger): type message responses and SSE schemas

* docs(swagger): type agent responses

* docs(swagger): type remaining handler responses

* docs(swagger): regenerate OpenAPI spec

* docs(swagger): type knowledge, activity and wiki maintenance endpoints

Covers the endpoints left out of the earlier passes: knowledge base
detail, document list/detail/spans/reparse/cancel/delete, knowledge
base activity and the Wiki stats, issues, lint and rebuild-links
endpoints. The spans handler now writes the documented structs instead
of gin.H, so the schema cannot drift from the body; the JSON is
unchanged. List filters gain their real enums and bounds, including
the finalizing, cancelled and deleting parse states.

* docs(swagger): document JSON-valued fields as JSON, not byte arrays

types.JSON and json.RawMessage are byte slices, so swag rendered every
metadata, details and config field as an array of integers. A swag
overrides file documents them as objects; the four fields that hold
arrays (relation chunk ids, allowed origins, image refs) say so with a
swaggertype tag.

* chore(lint): exempt swag annotation lines from the line-length limit

swag reads each @Param, @Success and @Failure as a single line, so an
enum list or a composed response type cannot be wrapped to fit.

* chore(lint): exempt struct tags carrying swag attributes from the line limit

Struct tags cannot wrap either; a field documented with swaggertype or
extensions keeps its tags on the declaration line.
2026-10-15 01:41:42 +08:00
lyingbug 383d26535a
feat(agent): persist human-in-the-loop prompts with a shared broker (#4250)
* feat(agent): persist human-in-the-loop prompts with a shared broker

Tool approvals and MCP OAuth prompts are now rows in agent_prompts,
awaited through hitl.Broker (lease, DB polling, Redis wake-up, per-kind
answer validation). Resolve is bound to session and kind. This is P1 of
the suspend/resume design; runs still block as before.

* feat(agent): suspend runs whose prompt outlives a short in-memory wait

Behind agent.prompt_suspend_enabled (off by default). A run raised from the
HTTP stream saves its state when a tool approval is raised; after the hot
wait the prompt moves to suspended and the assistant message is saved as
awaiting_input instead of blocking. A reaper takes over prompts whose
waiting replica stopped renewing its lease. Resume comes in a later change.

* feat(agent): resume suspended runs once their prompt is answered

The replica that records the answer resumes the run in process under the
prompt's lease. Superseded, expired and abandoned runs are closed with a
result for every tool call, and approved tools are never replayed.

* feat(agent): let users answer suspended tool approvals and follow the resumed run

* feat(agent): ask_user tool and generic prompt answer endpoint

- ask_user built-in tool: options, short forms or free text; validated
  answers, decline, suspend/resume like tool approval; non-interactive
  callers (IM, API keys) are told to ask in the reply instead
- POST /agent/prompts/{prompt_id} (and embed variant) answers ask_user
  and tool_approval prompts, returns resumed; SDK ResolveAgentPrompt
- agent_prompt_required/resolved events; AskUserCard sharing the
  countdown with ToolApprovalCard; answered questions rebuilt as cards
  from history
- awaiting-input messages keep their steps expanded so the pending card
  stays visible; prompt cards keyed by their own id
- i18n for all locales; design doc updated

* fix(agent): keep prompt answers that race a suspending or vanished owner

- Resolve suspends a pending prompt whose owner's lease lapsed, as the
  reaper would, so the answer resumes the run instead of waking nobody.
- Resuming waits for the suspending run to persist its message and clear
  its live-run marker; a marker left by a vanished owner is cleared.
- A resumed run that suspends again on a new prompt no longer has its
  message closed from the old state.
- tool_approval answers are decoded strictly and require "approved";
  fix the SDK doc that showed {"decision": "approve"}.
- API-key runs block instead of suspending, since they cannot answer.
- A run state that cannot be saved falls back to an in-memory wait, and
  NUL escapes are stripped from snapshots for jsonb.

* style(chat): align the ask_user card with the other interaction cards

Question and answer use the timeline's step text styles instead of
primary-colored and monospace text; options and form controls use the
compact step size, the number input spans the row, and the free-text
character counter is hidden.

* feat(chat): dock pending agent prompts above the composer

Tool approvals, ask_user questions and MCP OAuth prompts were answered
inline in the agent timeline, where a long answer or scrolling back
pushes them out of view. Open prompts of a streaming or suspended run
now gather in a dock above the composer (main chat and embed), one at a
time with a pager, lettered options and Enter to submit. The timeline
keeps a one-line record with a link back to the dock and the outcome
once answered. Views without a dock still answer inline.

* fix(agent): fence prompt ownership and preserve resumed turn context

* fix(agent): keep runs through consume errors and file prompts under the caller

Consume reported a store error as a lost claim, so an answer consumed
during a database hiccup detached its run: the message was never
persisted, the live-run marker stayed, and the reaper skipped the row
because the waiter had already dropped the snapshot. Consume now retries,
then decides from the row, and gives a run up only when the prompt is
known to have moved on.

Tool approvals and OAuth prompts of a shared agent were filed under the
agent's workspace, so the user's answer failed the tenant check and the
prompt timed out. Like ask_user, they now use the caller's workspace;
policy and token lookups stay in the agent's.

Enter on a picked option submits it instead of toggling it off, and
resume errors no longer report a nil cause.

* fix(agent): validate prompt choices and require durable approval claims

Reject malformed serialized option lists so the model can retry with valid arguments. Distinguish prompt consumption failures from lost ownership without authorizing side effects, and preserve native Enter selection for unselected choices.

* fix(agent): resolve prompt workflow lint errors
2026-10-14 20:37:28 +08:00
lyingbug 3b3b519310
fix(agent): clarify attachment image paths and chart readings (#4283) 2026-10-14 14:23:28 +08:00
Ochan Kwon e0712df3e0
feat(mcp): let callers without a browser obtain MCP OAuth tokens (#4229)
WeKnora runs behind other applications: they sign their users in and
call WeKnora with a tenant API key and X-External-User-ID, and WeKnora
keeps sessions and MCP OAuth tokens per external user. When an agent
uses an OAuth-protected MCP service, the only way to get that user's
token is the browser authorization code flow, and in this deployment
the user never sees WeKnora, so there is no browser to complete it.
Independently, an OAuth MCP service can only get its client through
dynamic client registration (RFC 7591), so authorization servers
without a registration endpoint cannot be used at all ("server does not
support dynamic client registration"), and servers that allow a grant
only to clients an administrator registered refuse a dynamic one.

JWT bearer grant. The application in front already knows who the user
is. POST /api/v1/mcp-services/:id/oauth/assertion lets it present a
signed assertion of that identity, such as the Identity Assertion JWT
Authorization Grant (ID-JAG) used by MCP's Enterprise-Managed
Authorization extension. WeKnora sends it to the service's
authorization server with the JWT bearer grant (RFC 7523) and stores
the token exactly like the callback does, so status, revoke, runtime
connections and, when the response carries a refresh token, refresh
use it unchanged. The assertion is forwarded unread; validating it is
up to the authorization server. The request reuses the principal from
authorize-url, the SSRF-safe HTTP client and the callback's connection
recycling, and the client the service already has. It does not run
dynamic client registration, which describes a redirect-based client;
without a client it answers 400. The token request outlives the caller,
so a token issued for an assertion that can only be used once is not
lost. A 4xx response other than 408 and 429, or a 2xx response carrying
an OAuth error, answers 400 with the OAuth error code and is logged; a
5xx, 408 or 429 response or a network failure answers 500. Request
logging masks the assertion field.

Pre-registered client. auth_config.client_id names a client registered
with the authorization server in advance, and client_secret is its
secret for a confidential client. When client_id is set it takes the
place of dynamic registration everywhere a client is read: the
authorization URL, the code exchange, the assertion grant and runtime
connections with their token refreshes. The values live on the service
rather than in mcp_oauth_clients, because a refresh rejected with
invalid_client deletes that row so the next authorization registers
again; an administrator's client would silently disappear. client_id is
non-secret: it is set through the main PUT, where an empty value clears
it, and returned in auth_config. client_secret follows api_key and
token: encrypted at rest, never returned, accepted in the create body,
and changed afterwards through the credentials subresource. Changing
client_id rebuilds the service's connections, and a connection built
with a pre-registered client is not treated as holding a replaced
registration when an authorization recycles connections. The MCP
service dialog adds optional Client ID and Client Secret fields to the
OAuth section, in all six locales.

Nothing changes for existing services: with client_id empty every
existing path reads the stored registration and registers dynamically
as before, and the authorization code flow, its state, the callback and
the dynamic registration request are untouched. The authorization
server HTTP client is moved into a function that both paths call.
2026-10-13 16:37:24 +08:00
lyingbug 194bffbb0a
feat(agent): read chat attachments on demand (#4203)
* feat(agent): read chat attachments on demand

Agent turns pasted query-selected attachment excerpts into the prompt before the model could act. List stored files as a manifest so the agent can page and search them through read_attachment.

* fix(agent): preserve attachment access and image reads

* fix(agent): bound attachment cleanup and image processing

* Remove standalone read_attachment tool

* Read attachments and images through read_file

Merge read_attachment into read_file (attachment:// paths, numbered
document images, staged-copy aliasing, query search), return sandbox
images as images, describe images for non-vision models inside the
tool call with bounded concurrency and model-chosen detail/focus,
render image reads in the UI, and record image metadata in Langfuse.

* fix(agent): make attachment pagination complete and bound image reads

* refactor(agent): separate attachment reads and cache image observations

* fix(attachment): read image batches concurrently and harden description cache

- ReadImages authorizes once and describes images in parallel
- start VLM deadlines after a slot is held; longer timeout for full reads
- detach shared description calls from the first caller's context
- skip rewriting unchanged sandbox projections
- strip the description cache from temporary document API responses
- render upload text before calling the vision model

* fix(attachment): number images in document order and polish attachment UI

- return resolved images in document order instead of reverse
- show attachment file names instead of attachment:// IDs in tool titles
- use the shared file icon and a lighter card for chat attachments

* fix(ci): renumber attachment migrations and satisfy lint

Main already took sqlite 000035 and versioned 000116, so the attachment
lookup indexes move to the next free numbers. Also wrap the ReadImages
signature and drop an unused test parameter.

* fix(ci): expect sqlite migrations to reach version 36
2026-10-13 14:37:17 +08:00
Ochan Kwon 809bda7b2f
fix(agent): make agent tool selection match the tools the runtime registers (#4211)
The tools a smart-reasoning agent may use were defined in three places
that never checked each other: the agent editor's checkboxes, the stored
allowed_tools, and what registerTools registers. The editor drew from a
hard-coded list that had drifted (search_conversations, a default tool,
had no checkbox), hid stored entries outside that list, and could not
untick a tool whose knowledge base capability was missing. The save API
stored any name. And an empty list could not mean "no tools": three
places treated it as "use the default tools", so clearing every checkbox
gave an agent without a knowledge base search_conversations, while the
editor's preview said it had none.

One source for the tool list: GET /api/v1/agents/tools serves the tools
of AvailableToolDefinitions with their group, danger flag and knowledge
base requirement, the default tools, and the tools other agent settings
register whatever the list says. The editor renders from it, falling
back to the catalog's label for a tool it has no translation for.

Stored values stay visible: a missing capability only blocks ticking a
tool, and stored names the catalog does not offer are shown ticked in
their own section and can be removed.

An empty list now means no built-in tools; only a list that was never
set selects the defaults. tools.EffectiveAllowedTools makes that
decision for tool registration, the session config and the rerank
requirement. Migration 000116 (SQLite 000035) writes the default tools
into the empty list of every existing smart-reasoning agent, so they
keep the tools they ran with. The knowledge base filter follows the
stored list, so such an agent with an "all" scope now gets the same
scope as one that ticked the defaults: bases search_knowledge cannot use
leave it. The editor writes the defaults into an unset list it loads,
so saving does not silently turn the defaults into no tools.

Saving validates the list: retired names are rewritten, switch-managed
and retired skill tools are dropped, names the request adds that no
tool answers to are rejected with a 400 that lists them, and unknown
names already stored are kept so an existing agent can still be saved.
The response reports what was kept or dropped.
2026-10-13 11:29:24 +08:00
tayadwx 98362dd296
feat(gallery): filter images by source document (#4159)
* feat(gallery): filter images by source document

Add a gallery toolbar button that opens a paginated, keyword-searchable
document picker; confirming restricts the listing to images owned by
the selected documents. The backend gallery endpoint accepts a
knowledge_ids query param (capped at 50) narrowing chunk_images by the
owning document before de-duplication, so the count query stays index
served.

* feat(gallery): lift knowledge_ids cap via POST body, sortable doc picker

Add POST /knowledge-bases/:id/images accepting the gallery filter as a
JSON body so knowledge_ids escapes the URL-length cap (50 -> 1000, an
abuse guard rather than a transport limit); the GET variant keeps its
cap and both share the contract-constrain logic. The source picker
gains sortable column headers (file name / upload time, newest first
by default), a header select-all that unions and removes whole hit
lists across keyword searches, overflow guards capping the no-keyword
view at the 500 most recent documents, and a wider dialog.

* style(gallery): replace td-token fallbacks and font-size literals with design tokens

styleGuard.test.mjs enforces ratchets: --td-* vars must not carry
fallbacks (theme.css guarantees them) and font sizes must use
--app-text-* tokens. GallerySourcePicker.vue added 4 of each,
pushing the ratchets over baseline.

* fix(gallery): keep shared images in source filter and align picker UI

- Scope the de-duplication probe to the selected documents, so an image
  whose newer copy lives in an unselected document is no longer hidden.
- Reset the source filter when the gallery switches knowledge bases.
- Accept an empty POST /images body; drop pruned attribute maps to nil.
- Keep the source filter out of the filter panel's count and clear action;
  the empty state's clear still resets it.
- Rework the picker dialog to the BatchTagDialog shell: heading with
  subtitle, footer with selection count, clickable rows, debounced search,
  select-all toggling against every hit, stale select-all responses dropped.
- Remove the unreachable GET knowledge_ids branch in the API client.

---------

Co-authored-by: tayadwx <tayadwx@users.noreply.github.com>
Co-authored-by: wizardchen <wizardchen@tencent.com>
2026-10-12 15:43:28 +08:00
JLi 643f64180e
feat(system): add platform user management and streamline settings (#4153)
* feat(system): platform user management — list, enable/disable, reset password, role toggle

Adds a SystemAdmin-only user management panel under platform settings
(System global -> Access). The backend already had repository primitives
and promote/revoke endpoints; this wires the missing service/handler/route
for listing and account-status changes and adds the UI.

Backend
- New GET /system/admin/users: paginated, cross-tenant user list
  (offset/limit, limit capped at 200), serialised via ToUserInfo so password
  hashes are never exposed; has_more derived with a limit+1 fetch (no COUNT).
- New POST /system/admin/users/status: enable/disable an account.
  Disabling revokes all of the user's tokens immediately. Service guards
  reject self-disable and disabling the last active system administrator.
- ValidateToken now re-checks is_active on every request, so a disabled
  account cannot keep using a previously issued access token.

Frontend
- UserListDialog: table with status/role tags, pagination, in-page
  username/email filter, enable/disable and reset-password actions, and
  one-click system-admin promote/revoke (reuses existing endpoints) with
  confirmation prompts. Reset-password dialog is a sibling (not nested)
  dialog so close/cancel/overlay all work.
- All new strings are i18n keys under system.globalSettings.userList in
  all six locales (en-US, zh-CN, zh-TW, ja-JP, ko-KR, ru-RU).

Tests: handler unit tests for list pagination/redaction/limit-cap and for
status-change success/guard/404 paths; existing handler/service/router
suites pass.

Co-authored-by: MOSS Agent <noreply@example.com>

* fix(system): make account status changes atomic and validate requests

* refactor(system): streamline settings and add dedicated user management tab

* fix(system): resolve user management CI lint findings

* fix(system): preserve account status during ordinary user updates

---------

Co-authored-by: desksk <26273255+desksk@users.noreply.github.com>
Co-authored-by: MOSS Agent <noreply@example.com>
Co-authored-by: wizardchen <wizardchen@tencent.com>
2026-10-11 17:07:00 +08:00
AstonKrss 0d39af1214 fix(retrieval): skip KB retrieval for summarize intent 2026-10-10 20:51:04 +08:00
hahahahahayesyeseys 89e3273daf fix(mcp): preserve disabled state on service creation 2026-10-10 20:50:05 +08:00
lyingbug c3f077ed18
fix(models): align configuration and multimodal embedding workflows (#4080)
* fix(models): align configuration defaults and multimodal workflows

* feat(image): enhance image retrieval diagnostics and filtering

- Implemented `retrieveObservedImagePool` to observe engine calls and record image recall decisions without altering results.
- Added detailed diagnostics in `summarizeImagePool` and `summarizeImageCandidateCut` to track image filtering decisions and scores.
- Introduced tests for image pool diagnostics to ensure accurate filtering and decision recording.
- Updated existing functions to utilize new retrieval and summarization methods for improved observability and accuracy in image handling.

* feat(embeddings): enhance image handling with new formats and message support

- Introduced `EmbeddingImageFormat` to manage image input formats, allowing for both object and messages formats.
- Updated `EmbeddingsSettings` to include new fields for `ImageFormat`, `ImagePrompt`, and `TextAsMessages` to support multimodal embedding workflows.
- Enhanced `Client` methods to build request bodies for images in both formats, ensuring compatibility with vLLM's requirements.
- Added tests to validate the new image message handling and ensure correct request formatting for different scenarios.
- Updated documentation to reflect changes in image input handling and configuration options.

* feat(similarity): add embedding similarity mode for enhanced document scoring

- Introduced a new `similarity` option in `ModelDebugOptions` to score documents and an optional image against a query.
- Updated `ModelDebugDrawer.vue` to support a new input mode for similarity, allowing users to select similarity as an input type.
- Enhanced the UI to display similarity results, including ranked candidates based on cosine similarity.
- Added tests to validate the functionality of the new similarity mode in embedding models.
- Updated localization files to include new strings related to similarity mode in multiple languages.

* feat(embeddings): add Qwen3-VL-Embedding support for multimodal input

- Introduced Qwen3-VL-Embedding model with support for both text and image inputs.
- Updated model configuration files to include compatibility settings for the new embedding type.
- Enhanced tests to validate the functionality of Qwen3-VL-Embedding, ensuring correct handling of inputs and outputs.
- Updated documentation to reflect the new model's capabilities and usage instructions.

* test(models): wrap embedding assertions to satisfy lint

* test(opensearch): isolate HTTP transports between parallel tests
2026-10-10 11:06:00 +08:00
lyingbug 65f88f4121
feat: judge image hits by their image in rerank and show them to vision models (#4013)
* feat(models): expose embedding input modalities

Whether an embedding model maps images into the same space as text is
what will decide if a knowledge base can store image vectors. The
catalog already had the field, but most multimodal embedding models
were missing it and embedding rows never reported it.

- Catalog: declare input modalities for the Doubao and Tongyi vision
  embeddings, the Aliyun multimodal / qwen-vl embedding families and
  jina-embeddings-v4, as listed in each vendor's documentation. The
  migration baseline changes for exactly these nine entries.
- Model responses resolve capabilities for embedding rows too, so
  capabilities.input reflects the catalog merged with spec.input.
- Model editor: an "Image input" switch for embedding models. It follows
  the catalog and writes spec.input only when the row differs from it;
  picking another catalog model or switching vendor drops the
  declaration. Model cards show the image badge for such models.

* feat(embedding): embed images with multimodal embedding models

Adds the image side of an embedder, for models that map images into the
same space as their text vectors so a text query can find an image.

- api: EmbedImage (bytes + MIME type, sent inline as base64 because the
  object store is usually unreachable from the vendor) and an
  ImageEmbedder implemented by all four protocols, each in the image
  shape its reference documents: Ark image_url, DashScope contents[].image,
  Gemini inlineData parts, and on the OpenAI shape an object keyed by the
  vendor's image_field (Jina: "image"). A plain OpenAI-compatible
  endpoint names no image field and takes no images.
- Settings: image_field, max_image_batch_size (default one per request),
  max_image_bytes and image_mime_types, filled from vendor docs for the
  DashScope multimodal models, gemini-embedding-2 and Jina.
- embedding: an optional ImageEmbedder interface reached through
  AsImageEmbedder, so existing Embedder implementations and fakes are
  untouched. An embedder takes images only when the catalog (or the
  row's spec.input) lists image AND the endpoint can carry one. Images
  are checked against the documented limits before any request, then
  split at the documented images per request. The concurrency, debug
  and Langfuse decorators forward it; logs and traces carry the MIME
  type and size, never the bytes.

The migration baseline changes only for the entries whose resolved
embedding settings now carry image limits.

* feat(knowledge): index images by their own vector and recall them

When the knowledge base's embedding model embeds images, each image now
gets a vector of its own next to its caption and OCR chunks, so a query
can find an image by what it shows even where the caption left it out.

Ingestion
- After the caption/OCR chunks are indexed, the multimodal task embeds
  the image (embedding.PrepareImage converts and shrinks it to the
  vendor's documented limits) and stores it as a new image_vector chunk:
  Content is the caption (OCR text when there is none) for reranking and
  the answer context; its index row carries the image's vector with
  SourceType ImageSourceType. Scanned PDF pages are skipped, their OCR
  text serves better. A failure here is logged and never fails the task.
- The image vector gets its own chunk rather than a second index row on
  the caption chunk: OpenSearch and Weaviate key documents by ChunkID,
  so a second row would overwrite the caption's text vector, and copy,
  move and enable/disable all follow chunk IDs.
- CompositeRetrieveEngine.BatchIndexVectors writes precomputed vectors
  through each engine's normal BatchIndex, keeping row order (the
  composite's de-duplication does not).
- Paths that rebuild vectors from text leave image_vector alone:
  syncChunkIndex only syncs its enabled flag, updateChunkVector skips
  it, and a caption edit updates its Content only.
- The multimodal task's caption/OCR index rows now set IsEnabled; only
  backends with a column default were storing them as enabled.

Retrieval
- A text query scores lower against an image than against text, so when
  the model embeds images the document vector search is widened by half
  (capped at 500) and run at min(VectorThreshold, 0.1); each store
  group's hits are then held to their own threshold before score
  normalization: text at VectorThreshold, images at min(VectorThreshold,
  0.1). Keyword hits on image rows are dropped, their Content being the
  already-indexed caption.
- image_vector hits are searchable and merge like caption hits: parent
  context is attached, and the caption is not repeated from ImageInfo.

Adds golang.org/x/image for scaling and WebP/BMP decoding.

* fix(models): report no thinking levels for embedding, rerank and ASR models

Embedding rows now carry capabilities, and the end-to-end run showed them
listing off/auto/minimal/low/medium/high: these references resolve no chat
protocol, and an empty level map reads as "every level supported". A model
without a chat protocol does not think, so its capabilities say so.

* fix(compaction): never read a cancelled turn's stream as a summary

streamSummary selects on the turn's context and the provider stream at
once. When the turn is already cancelled both are ready (providers close
the stream once their context ends), and select picks at random, so a
stopped turn could come back as a completed, partial summary with a nil
error. TestSummarizationIsNotRetriedOnceTheTurnIsCancelled caught it
intermittently in CI and reproduces under -race with several -cpu values.

Check the context before each receive and again when the stream closes.

* feat(rerank): score images with multimodal rerank models

Rerank models that read images can now be sent an image document and
score it against a text query on the same scale as text, which is what a
recall made by an image's own vector needs: a text reranker only sees the
caption, and rejects an image for what the caption left out.

- api: ImageReranker, implemented by the Cohere and NIM shapes. Image
  documents go in requests of their own, so text requests keep their
  exact shape. NIM sends image-only passages without text, as base64 data
  URLs (its VLM reranking reference requires both). The Cohere shape sends
  {"<image_field>": data URI} documents and asks for no echo of them.
- The image settings embeddings already had (image_field,
  max_image_batch_size, max_image_bytes, image_mime_types) move into a
  shared ImageInput embedded in both settings; resolved embedding settings
  serialize exactly as before.
- Catalog: jina-reranker-m0 and nvidia/llama-nemotron-rerank-vl-1b-v2 take
  text and image. Jina names "image" documents; its base64 form is not
  shown in any reranker example, so the data URI its embeddings reference
  documents is used and marked unverified. NVIDIA lists its six formats
  and 512 passages per request.
- rerank: an optional ImageReranker reached through AsImageReranker, gated
  on the model's catalog input and the endpoint; images are checked against
  the documented limits before any request, scores are normalized like
  text, and the debug and Langfuse decorators forward it without logging
  image bytes.
- Image preparation (format conversion, shrink to fit) moves from the
  embedding package to internal/models/imageprep, since rerank needs it
  too.

* feat(retrieval): keep a vector hit's own similarity through fusion

Hybrid fusion replaces a hit's Score with a rank-based RRF score. That is
right for ordering, but it erases the evidence reranking needs to judge an
image found by its own vector: how close the vector actually was. Record
the engine's similarity on each vector hit (IndexWithScore.VectorScore)
before normalization and fusion rewrite Score, and carry it onto the
search result (SearchResult.VectorScore, not serialized).

* feat(reranking): judge image hits by their image, not only their caption

An image found by its own vector was reranked on its caption and OCR text,
so a text reranker rejected it for exactly what the caption left out.

- With a reranker that reads images and a LoadImage callback, image_vector
  hits are scored by their image (prepared to the model's limits) on the
  same scale as text, and pass or fail the same threshold. A hit whose
  image cannot be read, prepared or scored falls back to its text.
- With a text reranker, its verdict stands for images that are mostly
  text (OCR of 200 characters or more). A pictorial hit it rejects is kept
  when its vector similarity (VectorScore) reaches Options.ImageKeepScore:
  at most two, strongest vector first, appended after the ranked results
  and marked kept_by=image_vector. This also returns an image when no text
  passes at all.
- Diagnostics report images_scored and images_kept.

* feat(rerank): read image hits' images in every rerank path

The chat pipeline, the agent search_knowledge tool and the hybrid-search
API now pass the rerank stage a way to read an image hit's image and an
image keep score, so a multimodal reranker scores images by what they show
and a text reranker can no longer drop a pictorial image its caption
undersells.

- KnowledgeBaseService.ReadChunkImage reads the first image of a result's
  image_info from the storage of its knowledge base (the knowledge base's
  tenant, so shared knowledge bases read from their owner's storage),
  capped at 32 MB. Plain http(s) images go through the SSRF-safe
  downloader.
- The backend resolution the image multimodal task used moves into a
  shared knowledgeImageReader, so ingestion and reranking find an image in
  the same place.
- reranking.ImageKeepScoreFor: an image is kept over a text reranker only
  when its vector was at least as close as a text hit had to be, and never
  below 0.25 (empirical).

* feat(chat): show a vision model the images its answer rests on

An image recalled by its own vector reached the answer as its caption only,
so a vision model could not answer from what the caption left out.

- Knowledge QA: INTO_CHAT_MESSAGE reads, for a model that supports vision,
  the images of the contexts that rest on an image vector match (at most
  three distinct ones, prepared to what every vision API takes: PNG, JPEG,
  WebP or GIF within 5 MB), attaches them after the user's own images, and
  names the contexts they belong to.
- Agent: search_knowledge attaches the same images when the agent's model
  supports vision (the agent describes tool images to models that cannot
  see them, and these already carry captions). Stored history drops them;
  they are read again from storage.
- An image hit and its caption hit read the same once merged, and
  de-duplication keeps either. The kept copy now inherits the image
  (image_vector_match), in the pipeline's exact and partial-overlap
  de-duplication and in the agent tool's.
- Results the rerank stage kept outside its top-k (kept_by) no longer fall
  to the later top-k cuts: the pipeline's FilterTopK and the agent tool's
  MMR and limit let them ride along after the ranked results.

* docs(website): describe how image hits are reranked and shown to vision models

* Fix image evidence preservation and model source mapping
2026-10-09 13:01:48 +08:00
tayadwx a1450d382e
fix(knowledge): claim cloned and released document images in the resource catalog (#3752)
* fix(knowledge): claim cloned and released document images in the catalog

#3493 binds images at parse time (bindStoredImages), which covers ingestion
but not the two paths that move image ownership:

Clone. A copied image is written as a new object, yet nothing claimed it for
the destination knowledge, so the KB file proxy answered 403 and cloned
documents rendered without images. The copy now goes through SaveBytes into
the servable exports/ namespace (CopyFile wrote to the knowledge-scoped
upload layout the proxy rejects), is claimed by the destination knowledge,
and CloneChunk / FAQ clone bind the handles left in chunk content and
image_info.

Delete. Release only looked at multimodal ImageInfo, which stays empty for
documents whose images live in markdown alone, so those objects outlived the
knowledge that owned them. The release set is now the union of ImageInfo URLs
and the bindings recorded for the knowledge being removed, with source_file
excluded so reparse and cleanup keep the original document.

The parse-path claim from the earlier attempt is dropped: ProcessDocument
already gets it from bindStoredImages, and the removed test only string
matched that call in the source. Every claim path stays best-effort,
including the clone path, which used to fail the whole clone and delete the
fresh copy when a single Bind errored.

* fix(knowledge): keep markdown-only images out of the pre-reindex cleanup

cleanupKnowledgeResources runs before a knowledge is re-indexed -- a manual
update, a reparse, a move with reparse -- not before it is removed. Unioning
in the catalog bindings there released a solely owned markdown-only
attachment, deleteExtractedImages deleted the file, and the catalog file
service marked the resource deleted. The re-claim that runs after the cleanup
resolves through a state=active lookup, so it could never bind that handle
again: the republished body was left pointing at a deleted image.

Cleanup collects ImageInfo URLs alone again. The paths that do remove a
knowledge entry (executeKnowledgeDelete, ProcessKBDelete) keep the union, which
is what lets markdown-only images be released when the knowledge actually goes
away.

Reported by @lyingbug in review on #3752.

---------

Co-authored-by: tayadwx <tayadwx@users.noreply.github.com>
2026-10-09 13:01:26 +08:00
linus-liu-web abc365f919
fix(wiki): 唯一写出口 UpdatePage 拦住丢表格行的机器重写(agent 整页写入不再静默改短页面) (#3793)
* fix(wiki): 在唯一写出口 UpdatePage 拦住丢表格行的机器重写

模型被要求把整页重新吐一遍时,markdown 表格是最常出错的地方:一百行的
证书台账回来只剩三十行,finish_reason=stop、没有任何报错,页面历史里就是
一次普通编辑。守卫因此加在 UpdatePage —— 所有调用方共用的唯一写出口,
而不是某一个调用方。

- 按「每行第一个非空单元格」的身份集合比较新旧内容(去掉强调符号、折叠
  空白、统一大小写):换行、重排、改日期/金额列都不算丢行,只有一行指向
  的实体整个消失才算;
- 同一身份出现两次按多重集比较,丢掉其中一行同样是丢;
- 表头行(`| --- |` 的上一行)不参与比较,模型改列名不算丢行。

有意的缩短放行:人在页面上删行(user)、回滚到更短的旧版本(revert),
以及调用方用 types.WithWikiShrinkAllowed 明确标注的写入 —— ingest 的
retract 轮(源文档被删除,被删文档名下的行本来就该消失)与 agent 的精确
文本替换(old_text 必须逐字命中现有内容,删除是有意为之)各自在自己的
写入上打这个标记。

agent 的整页写入拿到的是错误而不是静默成功:否则模型会以为改了一页、其实
一个字节都没动;它可以把那几行重新吐出来,或改用精确替换工具。

测试:身份比较、多重集、表头跳过、散文中孤立竖线不算表格;sqlite 上的
UpdatePage 拒绝写入且版本不推进、四类放行都能写进去、正常重写照常落库。
反向验证两组:去掉豁免 → 4 个放行用例 FAIL;让身份比较失效 → 6 个用例
FAIL(含两个 UpdatePage 用例);文件逐字节还原后全 PASS。

* fix(wiki): 拒写要向调用方明说,否则 trim 会删掉没落地的文档

UpdatePage 的守卫对非 agent 来源是**静默拒写**(返回现有页面 + nil
error)—— 页面本身没坏,这个选择是对的。但对 ingest 这条调用链来说,
"静默"本身就是个洞:

  reduceSlugUpdates 拿到 err == nil
    → 返回 changed=true, err=nil
    → 调用方那句 `if reduceErr != nil { collectUnapplied(updates) }` 走不到
    → trim 阶段认为这个 slug 已经成功合并
    → 把它背后的文档删掉
    → 而这些文档的贡献从来没写进页面,下一批也不会再回来

collectUnapplied 是"这些文档别删、下一批再来"的唯一入口,全仓只有三个
调用点(lockErr / !acquired / reduceErr != nil),没有一个能覆盖静默拒写。

修法不改 UpdatePage 的契约:拒写时它返回的是**未经改动的页面**,所以
调用方比较写入前后的 Version 就能判断,然后把 updateDeferred 一路带出去,
交给 collectUnapplied —— 与 #3617 的 updateDeferred 语义一致。

补的是 #3617 里这块语义;守卫本身仍在唯一写出口(不改架构)。
本提交只补连线,没有新增单测:这条路径要 asynq + chat model 的整套夹具,
而"静默拒写返回 nil"这一前提已由 TestUpdatePageRefusesARewriteThatDropsTableRows 覆盖。

* fix(wiki): 表格行守卫按去重身份集算比例,拒写改用显式信号

- agent 整页更新仍是任一身份缺失即拒写;ingest(未显式允许缩减)改为
  丢失超过旧内容唯一身份数 20% 才拒写,user/revert/显式允许缩减不变
- wikiWriteMissingRowIdentities 改为集合比较:同身份的两行合成一行不再
  算丢失;身份只取首个非空单元格,注释与日志都写明这个边界
- UpdatePage 的拒写统一返回 ErrWikiWriteDroppedTableRows;reduce 用
  errors.Is 区分"守卫拒写"与"成功但版本未变",不再拿版本号当信号,
  只有确实未落地的贡献才保留关联文档重试

---------

Co-authored-by: linus-liu-web <11640009+linus-liu-web@users.noreply.github.com>
Co-authored-by: linus-liu-web <linus-liu-web@users.noreply.github.com>
2026-10-09 10:55:28 +08:00
linus-liu-web cbfc0fed4a
fix(chat): 在模型上下文中标注 FILTER_TOP_K 的子集截断,避免把前 N 条当作全集作答 (#3758)
* fix(chat): 在模型上下文中标注 FILTER_TOP_K 的子集截断

FILTER_TOP_K 排序后直接 searchResult[:topK],模型只拿到 topK 条却按全集
语气作答(聚合类问题最明显:名单只被截取排名靠前的若干条,模型仍输出
"这些人是……"),属于完整但错误的答案。

本次只让截断对模型可见:真正丢结果时记录 RetrievalTruncation{Shown,
Candidates};prepareMessagesWithModelContext 把它作为 retrieval_shown /
retrieval_candidates 传给 modelcontext;annotateSearchNotes 在
<retrieval> 内追加 <subset shown="N" candidates="M">,说明这里只是排名
靠前的若干条、无法据此给出完整名单或精确计数。未截断时字段为 nil,不产
生任何标注。

提示落在 model context 视图(模型真正读到的那份)而不是
RenderedContexts:后者会被 prepareMessagesWithModelContext 整段替换成
<retrieval> 视图,写在那里的说明到不了模型。

不做日志:截断几乎每次检索都会触发,#3565 把 info 升 Warn 已被维护者以
"只会变噪音"否决;这里要解决的是模型看不到截断,不是运维看不到。

不动检索策略:不提高 topK、不改召回与重排,避免影响既有延迟和答案质量。

测试(夹具为合成数值):
- filter_top_k_test.go:截断时 Truncation={Shown:topK,Candidates:before},
  未截断(结果数 ≤ topK)时为 nil;
- modelcontext/omitted_results_test.go:带 retrieval_shown=5 /
  retrieval_candidates=48 时输出 <subset shown="5" candidates="48"> 且仍
  以 </retrieval> 结尾;shown==candidates 或计数缺失时不出现;
- references_test.go:Truncation 非 nil 时最终 model context 含
  <subset shown="3" candidates="48">,nil 时不含;另有端到端用例串起
  FILTER_TOP_K → INTO_CHAT_MESSAGE → prepareMessagesWithModelContext,确认
  提示到达模型消息而不是被替换掉。

* fix(chat): 截断提示改条件式并覆盖重排/合并的提前裁剪

评审意见(#3758):原提示无条件要求模型 "say the answer may be incomplete",
会让普通事实问答也带上免责声明;且重排阶段与 merge 回退已可能提前裁剪,
FILTER_TOP_K 不再截断时这条提示根本不生成。

- subset 提示改用评审给定的条件式措辞:只有问题在计数或要求完整列举时
  才说明结果范围,其他问题照常回答;N/M 填实际计数
- 明确 M 的语义:它是首次裁剪阶段(rerank / merge / filter_top_k)收到的
  候选片段数,不是全库匹配总数
- 重排阶段的阈值过滤、候选上限与 MMR top-k 裁剪,以及 merge 回退到
  RerankTopK 的裁剪,都经 ChatManage.RecordRetrievalCut 记录;首次裁剪为准,
  后续更窄的裁剪不覆盖候选池
- references 用实际渲染的知识片段数作为 N;补重排提前裁剪、merge 提前裁剪
  到模型消息的端到端回归测试

---------

Co-authored-by: linus-liu-web <11640009+linus-liu-web@users.noreply.github.com>
Co-authored-by: linus-liu-web <linus-liu-web@users.noreply.github.com>
2026-10-09 10:54:37 +08:00
hahahahahayesyeseys 419f17fc65
fix(wiki): resolve titles for unloaded backlinks (#3931)
* fix(wiki): resolve titles for unloaded backlinks

* fix(wiki): return backlink titles with page details

* docs(wiki): verify backlink titles in page detail schema
2026-10-09 10:50:52 +08:00
lyingbug 0c5d28abb7
feat: index and recall images with multimodal embedding models (#3683)
* feat(models): expose embedding input modalities

Whether an embedding model maps images into the same space as text is
what will decide if a knowledge base can store image vectors. The
catalog already had the field, but most multimodal embedding models
were missing it and embedding rows never reported it.

- Catalog: declare input modalities for the Doubao and Tongyi vision
  embeddings, the Aliyun multimodal / qwen-vl embedding families and
  jina-embeddings-v4, as listed in each vendor's documentation. The
  migration baseline changes for exactly these nine entries.
- Model responses resolve capabilities for embedding rows too, so
  capabilities.input reflects the catalog merged with spec.input.
- Model editor: an "Image input" switch for embedding models. It follows
  the catalog and writes spec.input only when the row differs from it;
  picking another catalog model or switching vendor drops the
  declaration. Model cards show the image badge for such models.

* feat(embedding): embed images with multimodal embedding models

Adds the image side of an embedder, for models that map images into the
same space as their text vectors so a text query can find an image.

- api: EmbedImage (bytes + MIME type, sent inline as base64 because the
  object store is usually unreachable from the vendor) and an
  ImageEmbedder implemented by all four protocols, each in the image
  shape its reference documents: Ark image_url, DashScope contents[].image,
  Gemini inlineData parts, and on the OpenAI shape an object keyed by the
  vendor's image_field (Jina: "image"). A plain OpenAI-compatible
  endpoint names no image field and takes no images.
- Settings: image_field, max_image_batch_size (default one per request),
  max_image_bytes and image_mime_types, filled from vendor docs for the
  DashScope multimodal models, gemini-embedding-2 and Jina.
- embedding: an optional ImageEmbedder interface reached through
  AsImageEmbedder, so existing Embedder implementations and fakes are
  untouched. An embedder takes images only when the catalog (or the
  row's spec.input) lists image AND the endpoint can carry one. Images
  are checked against the documented limits before any request, then
  split at the documented images per request. The concurrency, debug
  and Langfuse decorators forward it; logs and traces carry the MIME
  type and size, never the bytes.

The migration baseline changes only for the entries whose resolved
embedding settings now carry image limits.

* feat(knowledge): index images by their own vector and recall them

When the knowledge base's embedding model embeds images, each image now
gets a vector of its own next to its caption and OCR chunks, so a query
can find an image by what it shows even where the caption left it out.

Ingestion
- After the caption/OCR chunks are indexed, the multimodal task embeds
  the image (embedding.PrepareImage converts and shrinks it to the
  vendor's documented limits) and stores it as a new image_vector chunk:
  Content is the caption (OCR text when there is none) for reranking and
  the answer context; its index row carries the image's vector with
  SourceType ImageSourceType. Scanned PDF pages are skipped, their OCR
  text serves better. A failure here is logged and never fails the task.
- The image vector gets its own chunk rather than a second index row on
  the caption chunk: OpenSearch and Weaviate key documents by ChunkID,
  so a second row would overwrite the caption's text vector, and copy,
  move and enable/disable all follow chunk IDs.
- CompositeRetrieveEngine.BatchIndexVectors writes precomputed vectors
  through each engine's normal BatchIndex, keeping row order (the
  composite's de-duplication does not).
- Paths that rebuild vectors from text leave image_vector alone:
  syncChunkIndex only syncs its enabled flag, updateChunkVector skips
  it, and a caption edit updates its Content only.
- The multimodal task's caption/OCR index rows now set IsEnabled; only
  backends with a column default were storing them as enabled.

Retrieval
- A text query scores lower against an image than against text, so when
  the model embeds images the document vector search is widened by half
  (capped at 500) and run at min(VectorThreshold, 0.1); each store
  group's hits are then held to their own threshold before score
  normalization: text at VectorThreshold, images at min(VectorThreshold,
  0.1). Keyword hits on image rows are dropped, their Content being the
  already-indexed caption.
- image_vector hits are searchable and merge like caption hits: parent
  context is attached, and the caption is not repeated from ImageInfo.

Adds golang.org/x/image for scaling and WebP/BMP decoding.

* fix(models): report no thinking levels for embedding, rerank and ASR models

Embedding rows now carry capabilities, and the end-to-end run showed them
listing off/auto/minimal/low/medium/high: these references resolve no chat
protocol, and an empty level map reads as "every level supported". A model
without a chat protocol does not think, so its capabilities say so.

* fix(compaction): never read a cancelled turn's stream as a summary

streamSummary selects on the turn's context and the provider stream at
once. When the turn is already cancelled both are ready (providers close
the stream once their context ends), and select picks at random, so a
stopped turn could come back as a completed, partial summary with a nil
error. TestSummarizationIsNotRetriedOnceTheTurnIsCancelled caught it
intermittently in CI and reproduces under -race with several -cpu values.

Check the context before each receive and again when the stream closes.

* feat(knowledge): put image vectors behind a knowledge-base switch, off by default

Image vector indexing and recall used to turn on whenever the knowledge
base's embedding model declared image input. A model that takes images is
often chosen for text alone, so an upgrade would have added an embedding
call per image, widened vector recall 1.5x and lowered the image threshold
to 0.1 for those knowledge bases without anyone asking.

ImageProcessingConfig gains image_vector_enabled (default false; stored
JSON without the key reads as off, so existing knowledge bases stay off
with no migration). Both sides now need the switch and the model:

- indexImageVector returns before touching the model unless the KB opted
  in (KnowledgeBase.IsImageVectorEnabled, which also needs vector indexing).
- applyImageRecall resolves the embedding model only when some KB in the
  search opted in, and records the opted-in KBs per store group
  (storeGroup.ImageKBIDs). Only those groups widen the pool and lower the
  threshold; image hits of any other KB, including ones indexed before the
  switch was turned off, are dropped.

The KB editor shows the switch in the image processing section, with a
warning when the selected embedding model does not declare image input;
docs describe the switch, its cost and what turning it off does.

* fix(knowledge): leave image vectors behind when cloning into a KB that is off

CloneChunk copied every chunk of the source document and then CopyIndices
for all of them, so copying a document from a knowledge base with image
vectors on into one with the switch off (same model and store) still
created image_vector chunks and their index rows, bypassing the switch
that indexImageVector honours.

Image vector chunks are now dropped from the copy when the target has not
opted in. Every engine's CopyIndices copies only the rows of chunks in the
mapping, so their index rows stay behind too.

* fix(search): refill the vector pool past stale image vectors

Image hits of a knowledge base whose switch is off were dropped only after
the engine had ranked and cut its results at TopK. Image vectors indexed
before the switch was turned off could fill that pool and push out valid
text hits: with MatchCount 5 the pool is 50, and 50 stale image rows
ranked above the text left the search empty.

No engine can filter rows by source type (Postgres and SQLite do not even
take ExcludeChunkIDs), so the exclusion cannot move into the query. When
the document vector pool comes back full and holds stale image rows, that
search alone runs again on a pool twice as large, until it holds TopK rows
that are not stale, the index runs out, or the pool reaches
maxRetrievalPoolSize (at most four extra calls from a pool of 50). The pool
is then cut back to TopK such rows, matching an engine-side exclusion
unless more than maxRetrievalPoolSize-TopK stale rows rank above the text.

* feat(i18n): translate the image vector switch and image input keys into zh-TW

The zh-TW locale arrived from main after this branch added five keys: the
knowledge base's image vector switch (label, description, unsupported-model
warning) and the embedding model's image input flag (label, description).

* fix(knowledge): leave image vectors behind when moving into a KB that is off

A reuse_vectors move relocates a document's chunks and index rows in place:
MoveKnowledgeIndices rebinds the rows by document ID and
MoveChunksByKnowledgeID rebinds the chunks. Moving a document from a
knowledge base with image vectors on into one with the switch off therefore
carried its image_vector chunks and their vectors along, the same bypass the
clone path had.

When the target has not opted in, the image_vector chunks are now deleted
during the move, their index rows first and then the chunks, so a retry
finds the chunks again rather than leaving orphaned rows. A target that
opted in still receives them unchanged.

* fix(retriever): keep image vectors out of keyword-only engines

BatchIndexVectors handed image rows to every engine of the composite,
including one that serves keyword retrieval alone (Elasticsearch v7 next to
a vector store). There the row holds nothing but the image's caption, which
the caption's own chunk already indexes, and retrieval discards every
keyword hit on it.

Image rows now go only to engines that serve vector retrieval. An engine
that serves both keeps one row for both kinds, so its keyword search still
meets them; the search side handles that.

* fix(search): refill the keyword pool past image rows

Every engine that serves both kinds of retrieval stores one row per index
entry and runs its keyword search over that row's content, so an image
vector row, whose content is its caption, is a keyword hit wherever the
caption's own chunk is. filterImageHits drops all of them, but only after
the engine has cut its ranking at TopK, so they crowded out text and caption
hits; a keyword ranking that put the image rows first came back empty.

The keyword pool is now refilled the way the vector pool is refilled past
stale image vectors: when it came back full and held image rows, the
keyword search runs again on a pool twice as large, up to
maxRetrievalPoolSize, and is cut back to TopK hits that are kept.

Leaving the rows out of the keyword index instead would mean writing them
without searchable content (Milvus and Tencent VectorDB derive a BM25
sparse vector from it, and how they take an empty one is unverified) or a
source-type filter in every engine's keyword query (ten engines, and
filterable fields that existing collections lack).

* revert: let image vectors move with the document into a KB that is off

This reverts commit e259977fe915f7d0bc30c4b02bc7af04da1c8dd4.

Deleting the image vector index rows by chunk ID before the move is not
reliable: the Weaviate and Elasticsearch v8 delete wrappers ignore partial
failures, and OpenSearch rejects more than 1000 IDs with ErrBatchTooLarge.
A partial delete followed by the chunk delete leaves index rows whose chunks
are gone, which nothing ever cleans up.

Without it, the image_vector chunks and their index rows move together
with the document. While the target's switch is off they are filtered at
query time (the pool refill keeps them from crowding out text), they are
removed with the document on delete, reparse or knowledge base delete, both
chunks and index rows by knowledge ID, and they become active again if the
target turns the switch on. The clone path keeps dropping them, since there
the rows are copied rather than moved.

* fix(search): keep the image row refill from losing hits it already had

refillPool replaced a pool with the answer to each larger search. A refill
that answered only in part (RetrieveResult.Error, swallowed into a warning
by retainPartialResults, as when some collections of the store did not
answer) could hold fewer text rows than the complete smaller pool, or none:
40 stale image rows above 10 text came back from a partial refill as the
40 image rows alone, and the search lost all 10 text hits. A refill that
failed outright failed the whole store group, discarding a first round that
had succeeded.

A refill is now best effort. A failed or incomplete refill search stops
the refill and leaves the last complete pool in place, cut back to TopK
kept rows as before, so a refill can no longer leave a search worse off
than it was without it. Merging the partial answer into the pool was not
done: its rows would need deduplicating and its ranking is cut at a
different TopK over a different set of collections, while keeping the
previous pool is exact.

* docs(knowledge): say what happens to image vectors in a KB that is off and what the refill costs

The image vector section now says that with the switch off, image vectors
that were indexed before it was turned off, or that came along when the
document was moved in from a KB that has it on, are kept but filtered at
query time, are removed when the document is deleted or reparsed or the
KB is deleted, and are recalled again once the switch is turned on; a copy
into such a KB leaves them behind.

It also states the cost of keeping dropped image rows from crowding text
out of the pools: the vector and keyword pools are searched again with
TopK doubled up to the 500 candidate cap, which adds queries for store
groups that hold such rows, and text can still be missed when more than
about 500 candidates are image rows. A failed or partial refill keeps the
previous complete pool.

* fix(search): log unexpected refill result sets, and describe the refill stop condition precisely
2026-10-08 18:19:53 +08:00
Junhao Ye d2bc0df6a2
feat(i18n): add Traditional Chinese (zh-TW) locale support (#3916)
* i18n(zh-TW): add Traditional Chinese (Taiwan) locale bundle (6,547 keys)

機械產生自 zh-CN.ts:以 zhconv (zh-tw) 為字形基底,詞彙層以 tw-terms.tsv(364 條生效詞條)覆寫,保留 26 條 never 與 47 條保護詞。骨架等價、鍵序列、Node 真實載入、內嵌區塊 四道檢查皆通過(詳見審計目錄)。

* i18n(zh-TW): register locale bundle in i18n infrastructure (5 → 6)

- resolveDefaultLocale.ts: SUPPORTED_LOCALES = ['zh-CN','zh-TW','en-US','ru-RU','ko-KR','ja-JP']
- i18n/index.ts: 載入 zhTW 並掛入 messages['zh-TW']
- localeKeyAudit.ts: 加入 LOCALE_BUNDLES['zh-TW'] 與測試守門(LOCALE_ORDER、emptyBundles)
- localeKeyAudit.test.ts: 把 zh-TW 補進期望清單
- locales/{en-US,ja-JP,ko-KR,ru-RU,zh-CN}.ts: language 區塊補 zhTW 顯示名(讓語言切換器正確顯示繁中選項)

* i18n(zh-TW): fix normalizeEmbedLocale and add zh-TW embed block (514 keys)

原實作 startsWith('zh') 直接命中 zh-TW/zh-HK/zh-Hant 並回傳 zh-CN
→ 繁中使用者將永遠看到簡體字串。修正後:
  - 繁中變體 (zh-tw/zh-hk/zh-mo/zh-hant) 排在通用 zh 之前,優先回傳 zh-TW
  - 內嵌 zh-TW 區塊加入 514 個字串,骨架與 zh-CN 區塊逐字元相同(只換區塊名)

* i18n(zh-TW): wire zh-TW into UI language pickers and TDesign locale

- App.vue: 載入 tdesign-vue-next esm/zh_TW,tdLocaleMap['zh-TW'] = zhTWConfig
- Login.vue: 語言切換器新增 { value: 'zh-TW', label: '繁體中文', shortLabel: '繁中', flag: '繁' }
  (刻意不用旗標 emoji,改以「繁」字標示,語意清楚且無爭議)
- GeneralSettings.vue: 設定頁語言下拉新增 <t-option value='zh-TW' ...>
- AgentEmbedChannelPanel.vue: 代理嵌入頻道語言選單新增繁中
- IMChannelPanel.vue: IM 頻道語言選單新增繁中

* i18n(zh-TW): widen API union types and update tests for zh-TW

- api/agent/index.ts: IMChannel.locale 聯合型別補 'zh-TW'
- api/embed/index.ts: EmbedLocaleTag 聯合型別補 'zh-TW'
- i18n/embedLocale.test.ts: supported-list 期望補上 zh-TW;新增 7 條 normalizeEmbedLocale 斷言(覆蓋 zh-tw / zh-hk / zh-mo / zh-hant / zh-Hant-TW 等變體正確路由到 zh-TW)
- locales/workspaceTerminology.test.ts: workspace 詞彙測試也對 zh-TW 做同樣的禁用詞斷言(與其他 locale 同步)

* i18n(zh-TW): add zh-TW to backend supportedLocales whitelist

- internal/types/locale.go: supportedLocales["zh-TW"] = {}
  (NormalizeSupportedLocale 會靜默丟掉未列出的 tag,導致前端語言選擇器實際寫入空字串——
   此修正保證前端選 zh-TW 時 embed 預設語言與 IM 回覆語言會被後端接受)
- internal/types/locale_test.go: 期望清單補上 zh-TW
- internal/handler/mcp_usage_instructions.go: MCP 工具使用說明對照表新增 zh-TW → 'Traditional Chinese'

* fix(i18n): translate revoke as 撤銷 in zh-TW and allow zh-TW as the runtime default locale

The zh-CN → zh-TW conversion turned every 撤销 into 復原 (restore), so
revoke buttons for OAuth tokens, device pairing, invitations and system
admins read as the opposite action. Only the undo sense (此操作不可撤销)
keeps 復原.

docker-entrypoint.sh whitelists DEFAULT_LOCALE and did not accept zh-TW,
so DEFAULT_LOCALE=zh-TW silently fell back to Simplified Chinese. Also
list zh-TW in the IM locale error and the locale hints.

---------

Co-authored-by: junhao-ye <junhao-ye@users.noreply.github.com>
Co-authored-by: wizardchen <wizardchen@tencent.com>
2026-10-08 16:18:17 +08:00
zaqqmm 19542f2a1d
feat(datasource): add Seafile data source connector (#3616)
* fix(datasource): omit last_sync_cursor from the data source list response

The settings page polls GET /datasource every few seconds during a sync,
and connector cursors can grow to megabytes. No list consumer reads the
cursor, so NewDataSourceResponses now blanks it; detail, create and
update responses still return it through NewDataSourceResponse.

* feat(datasource): add Seafile connector with fingerprint-based incremental sync

Sync folders and files from one Seafile library per data source. The
connector walks the selected scope one directory level at a time,
downloads files through the link + fileserver flow without credentials,
and reconciles deletions against a per-file oid/mtime/size fingerprint
cursor that is checkpointed every 50 mutations.

- resource ids are "<repo_id>:<path>"; ancestors collapse, one library
- per-file failures become coded error items and keep the path pending
- directory failures and Retry-After > 60s abort without deletions
- forced full runs keep the old cursor as a deletion baseline
- API token never reaches cursors, metadata, logs or error text
- share the supported import extension set through internal/utils

* feat(datasource): add Seafile frontend connector and docs

Wire the Seafile data source into the editor dialog, channel labels
and the five locales, and document how to connect a Seafile 10.0.1
library.

- connector definition with base_url + API token fields and icon
- picker guard: one library per data source, at least one selection
- "library" resource type and icon in the lazy tree
- channel label in the knowledge base filter, document views and chat
- i18n for labels, guide steps and the eight seafile_* sync error codes
- website-docs integration guide, data source and extension point doc
  updates, and Seafile in the README data source lists

---------

Co-authored-by: zaqqmm <q-zhang10@qq.com>
2026-10-08 16:18:09 +08:00
5quan 3fb6b9d631
fix(kb): let the update API set vlm_config (#3704)
* fix(kb): let update API set vlm_config

UpdateKnowledgeBaseRequest only carried name/description/config, so the
PUT /knowledge-bases/:id endpoint could never change a knowledge base's
multimodal (VLM) config, even though CreateKnowledgeBase accepts a
top-level vlm_config. Thread an optional vlm_config through the handler
and service so the update path mirrors create (nil = no change).

Fixes #1723

* test(kb): pass the vlm_config argument in tests added on main

* fix(kb): accept only model-managed vlm_config fields on update

The update API copied the request's VLMConfig wholesale, so a caller
could set the legacy ModelName/BaseURL/APIKey/InterfaceType fields and
have the VLM client call BaseURL directly, outside model management and
its SSRF checks. Only Enabled/ModelID/DescriptionLanguage/
CustomInstructions are taken now; legacy fields keep their stored value.
A referenced model must exist and be a VLLM model (400 otherwise), and
disabling clears ModelID, matching the initialization config endpoint.

* fix(kb): validate and gate vlm_config on the update API

- Enforce the 4000-character custom_instructions cap
  (types.ValidateKnowledgeBasePromptInstructions), as create and
  /initialization/config already do.
- vlm_config is a KB setting: require kbSettingsAccess (owner workspace
  or admin share) like /initialization/config, so share editors get 403.
- Stored legacy model_name/base_url keep IsEnabled() true on their own,
  so disabling returned 200 but left VLM on, and managed -> disabled fell
  back to the old endpoint. Clear them when the request enables a managed
  model_id, or disables VLM without echoing back the stored
  model_name+base_url (GET-then-PUT keeps legacy on). Request legacy
  fields are only compared, never written. Enabling without model_id on a
  non-legacy KB is now a 400 instead of a silent no-op.
- Document the vlm_config semantics in the KB API reference.

* fix(kb): keep the legacy VLM endpoint only for pure legacy knowledge bases

CreateKnowledgeBase stores vlm_config as sent, so a managed model_id can
sit next to legacy model_name/base_url; resolveVLM then uses the managed
model. The update's keep-legacy paths (disable echoing the stored legacy
fields, enabled=true without model_id) cleared model_id but kept the
legacy fields, so IsEnabled() stayed true and image calls moved to the
stored base_url.

Those paths now apply only when no model_id is stored. On a mixed config
a disable clears everything, and enabled=true without model_id is a 400,
the same as on any managed knowledge base.

---------

Co-authored-by: wizardchen <wizardchen@tencent.com>
2026-10-08 14:37:23 +08:00
linus-liu-web 626185a120
fix(wiki): 重建链接逐页写失败不再静默,半成品不再回 200 (#3884)
RebuildLinks 的保存循环把每页 UpdateMeta 的失败降级成一条 Warn,函数最后
无条件 return nil,于是 handler 里既有的 err != nil 分支永远进不去:任意
一页写失败(行锁等待、死锁、连接中断、超时)时调用方仍拿到
200 "Links rebuilt successfully",而这些页面的 in_links/out_links 保持旧
值,反向链接与图谱继续显示过期关系,且没有任何途径知道"只重建了一部分"。

保持尽力而为的语义不变(一页失败不影响其余页面继续写入),但把失败页聚合
起来在结尾返回,沿用同仓库 knowledge_delete.go / wiki_ingest.go 的
errors.Join 写法,以及同文件 MovePage 对同一个 UpdateMeta 调用的报错写法。
handler 无需改动:其既有 500 分支会如实返回 "N of M pages failed" 与失败页
slug。

回归用例用 SQLite 触发器让指定页的 UPDATE 真的报错(不是 mock),走真实仓库
与真实 handler:
- 部分失败:RebuildLinks 返回聚合错误,可写页仍被刷新,失败页保持旧值;
- 全部失败:错误里含 3 of 3 与全部 slug;
- HTTP 层:部分失败回 500 且错误体含失败页与底层原因,全部成功才回 200。

Co-authored-by: linus-liu-web <11640009+linus-liu-web@users.noreply.github.com>
2026-10-08 01:25:53 +08:00
knqiufan e29301115a
fix(graph): preserve relationship direction and document provenance (#3950)
* fix(graph): preserve direction and document provenance in graph retrieval

* test(graph): verify live query bounds and seed evidence ordering

* fix(graph): deduplicate expansion using bounded seed identities

* ci: retry upstream scan after missing CodeCC report
2026-10-08 00:48:51 +08:00
Mohamed Arbi 9fa9e7c29d
fix: use consistent qdrant env settings (#3960)
* fix qdrant env settings

* fix test formatting
2026-10-07 23:59:56 +08:00
linus-liu-web 5f5093b9f7
fix(retriever): Qdrant/Milvus 关键词检索整批失败返回 error,不再当成「没有匹配」 (#3847)
* fix(retriever): Qdrant/Milvus 关键词检索整批失败不再当成没有匹配

Qdrant 与 Milvus 的 KeywordsRetrieve 逐个 collection 检索,单个 collection 失败只降级成
一条日志后 continue。循环结束后按结果条数分支:全部 collection 失败与确实零命中都会打印
"No keyword matches found"、返回空结果与 nil error。上层只看第二个返回值,据此判定知识库
没有相关内容,用户与模型都无法区分"没有这个词"和"检索没有执行成功"。

改为在循环内累计匹配到的 collection 数与失败的 collection 数,全部匹配 collection 都失败
时返回带底层原因的聚合 error,例如
"qdrant keyword search failed in all 2 matched collections: <cause>"。计数放在 collection
名前缀/维度匹配之后,名字不匹配的 collection 不参与统计;匹配到的 collection 为 0 时维持
原行为,不报错。Milvus 的三处 continue(Search / convertResultSet /
buildMilvusIndexResults)全部计入,另有两处同样会跳过 collection 的失败路径(analyzer mode
探测、过滤器构建)一并计入,这三处的日志补上 collection 名。

部分失败时不再打印结论性的 "No keyword matches found":有结果时改为提示结果不完整,无结果
时说明有多少个 collection 应答过、多少个失败;只有本次调用没有发生任何错误时才继续打印该
结论。RetrieveResult.Error 保持原样(仓库内没有读取方),改动面最小、语义最清晰的暴露方式
是整批失败返回 error。

回归测试:
- Qdrant(repository_test.go):假 Qdrant gRPC 服务让 Scroll 全失败时返回 error 且不打印
  结论日志;只有一个 collection 失败时结果与 nil error 不变;全部成功且零命中时仍返回空结果
  + nil error 并保留结论日志。
- Milvus(新增 keywords_test.go):*client.Client 没有接口缝隙,改用进程内 Milvus gRPC 假
  服务(ShowCollections / DescribeCollection / Search)覆盖同样三种情形,走真实 Search 与
  结果转换路径。
- go.mod:milvus-proto 由新测试直接导入,去掉该依赖的 // indirect 标记。

反向验证:把两个 repository.go 改回原样后,两条"全部失败"用例分别以
"all-collections failure must surface as an error, got <nil>" 与
"An error is expected but got nil" 失败,部分失败与零命中两条用例保持通过;随后逐字节还原
修复。

* fix(retriever): 关键词检索部分失败随结果一起暴露,不再被当成完整结果

Qdrant/Milvus 的 KeywordsRetrieve 逐 collection 检索,部分 collection 失败时保留其余
collection 的结果并返回 nil error。上层拿到的是"结果 + nil error":既不知道少了几个
collection 的内容,也没有任何读取方,RetrieveResult.Error 因此长期是死字段。上一个
commit 只处理了"全部匹配 collection 都失败",部分失败仍然静默降级,模型与用户看到的
都是不完整结果。

让部分失败沿着结果走完这条链路:

- repository(qdrant/milvus):部分失败时把
  "… keyword search failed in N of M matched collections: <cause>" 写进
  RetrieveResult.Error —— 结果照常返回、调用级 error 仍为 nil;全部失败仍是调用级
  error,真正的零命中仍然 Error == nil;
- CompositeRetrieveEngine:读取每个结果集的 Error 并聚合,连同结果一起返回
  ("retrieve returned partial results: …");引擎级 error 仍然丢弃全部结果,
  all-or-nothing 策略不变;
- retrieveFromStores(该接口的唯一消费方):结果与 error 同时非空即"部分失败",打出
  带 store group 与原因的 WARN 并保留结果返回 —— 既不会把降级 store 升级成整次搜索
  失败,也不会让上层把不完整结果当完整结果;
- matched == 0(没有任何 collection 匹配 base name)单独打一条 Warn,与"检索成功但
  零命中"区分开;
- milvus/repository.go 的 allFields 块按 gofumpt 重排,四个包 gofumpt -l 干净。

回归测试:

- qdrant/milvus:部分失败(有命中 / 零命中)结果集带 Error 且调用 error 为 nil;全部
  成功零命中仍 Error == nil 且保留结论日志;全部失败仍是 error;matched == 0 打专属
  Warn 且不发任何 Scroll/Search;
- CompositeRetrieveEngine:结果集带 Error → (结果, error) 同时返回;结果集完整(含零
  命中)→ 无 error;引擎级 error → 丢弃全部结果;
- retrieveFromStores:单 store / 多 store 部分失败都打 WARN 且保留结果,完整零命中不
  打 partial 日志,引擎级 error 仍然失败。

反向验证:去掉四个源文件的可见性改动后,17 条新/改用例中 9 条失败(qdrant 3、milvus 3、
composite 1、retrieveFromStores 2),钉住"行为不变"的 8 条保持通过;还原后 diff 与
改动前逐字节一致。

---------

Co-authored-by: linus-liu-web <11640009+linus-liu-web@users.noreply.github.com>
2026-10-07 23:59:38 +08:00
hailongzhao bccb4b151b fix(embed): 内嵌网页只传图片不输入文字时不再返回 400
内嵌网页的输入框允许只带图片或附件就点击发送,但 CreateKnowledgeQARequest.Query
带有 binding:"required",parseQARequest 也拒绝空 query,于是只传图片直接返回
400 "Query content cannot be empty"。

入口处理:去掉 binding:"required";文字为空但带有内联图片数据或内联附件时,
用 types.UploadOnlyQuestion 生成一句替用户提问的问题(中文界面为「请根据我
上传的内容回答。」,其他语言为英文),交给模型、检索、标题、会话历史索引、
追问建议和记忆使用。只有 URL 的图片不算上传,因为客户端传入的图片 URL 会被
清掉;预上传的 attachment_ids 也不算,这类文件在流开始后才解析,可能失败或
超时,届时模型没有任何内容可答。其余空 query 仍返回 400。

存储与显示:qaRequestContext 新增 userInput,保存用户消息时只存用户实际
输入,只传图片时为空,刷新后与发送当下显示一致;query 仍是给模型的问题。
steer 追问复制上一轮的请求上下文,显式设置 userInput,避免在只传图片的一轮
之后把追问存成空消息。

会话历史:文字为空但带图片或附件的用户消息,在两处历史重建里补上同一句
问题。知识问答流水线(loadAndProcessHistory)原先会整轮丢弃;Agent 历史
(LoadAgentHistory)原先会发出空的用户消息,被 SanitizeMessages 剔除后
前后两条回答被合并。

去掉 binding 标签会让 gofmt 重新对齐整个 CreateKnowledgeQARequest 的行尾
注释,这些既有的超长行因此会被 PR 的增量 lint 视为新增。按仓库惯例把字段
注释移到字段上一行(注释文字不变,swagger 描述不受影响),并把 Go 字段
KnowledgeIds 改名为 KnowledgeIDs(JSON 名仍是 knowledge_ids,接口不变)。

同步更新 swagger 文档,query 不再是必填字段。
2026-09-30 15:26:51 +08:00
LL_Ricky 9842e51d40
fix(frontend): 修复复制模型时模型名称被改写的问题 (#3830)
* fix(frontend): 修复复制模型时模型名称被改写的问题

* fix(model): preserve model name and credentials when copying a config

Copying rewrote the upstream model name and posted redacted parameters, so the duplicate could not authenticate. Clone the stored row, cap the display name, and skip a new download when the local model is already active.

* fix(lint): clear new golangci-lint findings on model copy

Check c.Error results, wrap the long test line, and format the stub. The pull request lint range also includes the new HTTP error test, whose unused request parameter failed revive.

---------

Co-authored-by: wizardchen <wizardchen@tencent.com>
2026-09-29 19:32:08 +08:00
Fei.Chen ec0ca46334
fix(server): run ResourceCleaner on signal shutdown instead of racing Fatalf exit (#3804)
* fix(server): signal shutdown raced Fatalf exit before ResourceCleaner ran

The signal goroutine closed the listener before calling server.Shutdown,
so Serve returned the raw accept error ('use of closed network connection')
instead of ErrServerClosed. The invoke error path then hit logger.Fatalf
(os.Exit(1)) in the same millisecond, killing the process before the
shutdown goroutine reached resourceCleaner.Cleanup. Every registered
cleaner — BrowserSkill daemon, sandbox reapers, IMService, AntsPool,
Langfuse — was skipped on every SIGTERM/SIGINT, leaking the shared bsk
daemon process and its temp home on each restart.

Two changes:
- Drop the manual listener.Close(): Shutdown closes all listeners as its
  first step, releasing the port just as fast without racing Serve.
- Give resourceCleaner.Cleanup its own fresh timeout context so a slow
  connection drain cannot leave Cleanup with an already-expired context,
  which made it skip every entry without running any.

* fix(server): keep signal cleanup inside one grace period

Desktop quit still closed the listener before Shutdown and returned
before ResourceCleaner finished, so Fatalf could skip every hook.
Share ShutdownTimeout between drain and cleanup, run BrowserSkill
first, and do not drop later hooks when the deadline expires.

* fix(server): satisfy errcheck and line-length lint

---------

Co-authored-by: vicfei <vicfei@users.noreply.github.com>
Co-authored-by: wizardchen <wizardchen@tencent.com>
2026-09-28 23:37:36 +08:00
wizardchen c819abd958 fix(wiki): bound outstanding follow-up tasks 2026-09-28 22:02:07 +08:00
lyingbug 9114e4e4f9
fix(citations): 修复原文定位、高亮范围与 DocReader 启动失败 (#3794)
* fix(citations): verify source locations and recover document previews

* chore: exclude local research notes from pull request
2026-09-27 20:32:26 +08:00