Available vs uninstalled skills were easy to mix up, and installing rows
lost their action. Group the list, keep a progress control, and open the
install drawer without leaving the agent editor.
E2B refuses to delete a skill template while paused session sandboxes
still reference it, but reports that as HTTP 400. Classify it as
Conflict, leave the ledger row for the next sweep, and do not mark the
Langfuse span ERROR — this is expected until those sandboxes expire.
Docker Engine HTTP was parenting to whatever agent.round was still
recording, and the idle-sweep goroutine leaked List/Stat/Delete into
the chat tree after the tool span ended. Wrap RemoteSandboxClient with
sandbox.* spans and clone background work without the chat trace.
POST /suggestions often wins the race to generate follow-ups and has no
Langfuse parent, so chat.completion auto-creates an orphan root. Stamp
the originating W3C traceparent on the assistant message and resume it
before the extra LLM call.
* fix(docreader): keep DOCX table content in parsed text
Docx.__call__ returns tables separately from the text lines, and
parse_into_text used to drop them, silently losing all table content
in documents parsed with the builtin engine. Render the tables as GFM
Markdown (header row + delimiter + data rows) and append them to the
parsed text.
Fixes#2881
* fix(docreader): render DOCX tables from grid cells in document order
Skip the HTML round-trip that treated equal or empty adjacent cells as
colspan and dropped columns from GFM. Emit one column per python-docx
grid slot and interleave tables with paragraphs in body order.
---------
Co-authored-by: root <root@localhost.localdomain>
Co-authored-by: wizardchen <wizardchen@tencent.com>
* fix(docreader): handle Header object when stripping mhtml header
* fix(docreader): decode 8-bit MHTML headers without crashing (#2901)
compat32 wraps non-ASCII 8-bit headers as Header; str() turns them into
U+FFFD and HTML Content-Location still crashed on .lower()/urljoin.
Refs: https://github.com/Tencent/WeKnora/issues/2901
---------
Co-authored-by: wizardchen <wizardchen@tencent.com>
* fix(dsh-weknora): show knowledge-base figures in tool results (#2829)
dsh cannot load resource:// handles, and the plugin previously dropped
image_info, so Q&A returned text only. Default to public URLs and emit
Markdown images so the harness can render cited figures.
Refs: https://github.com/Tencent/WeKnora/issues/2829
* fix(dsh-weknora): rely on public URLs in content instead of image_info
Passage text already carries Markdown figures after WeKnora rewrites
resource:// handles. Drop the extra images arrays and clip restoration.
ppt/pptx still fall back to markitdown when anydoc is not in the binary.
pdf/docx/xlsx and other anydoc formats now default to the in-process
converter as well; csv/md stay on the Go simple reader.
Type-level defaults still fall back to markitdown, but a linked anydoc
build should use the in-process converter first instead of always
shipping Office decks through docreader.
Empty parser_engine_rules used to resolve to builtin, which historically
did not register ppt/pptx and failed with an unusable ValueError. Route
those types to markitdown by default on both the Go resolver and the
docreader registry, and list engines that actually support the type.
Refs: https://github.com/Tencent/WeKnora/issues/2882
The Go client URL test was a false positive: missing WebKit used to
surface as markdown content. #2883 raises instead, so CI must install
the same browser the DocReader image uses.
Empty scrape/parse results were returned as "Error parsing web page: …"
and then chunked into the knowledge index, hiding the failure. Raise
WebParseError so DocReader returns error and the knowledge is marked failed.
Refs: https://github.com/Tencent/WeKnora/issues/2883
Streaming only cached the first fence, and the loading-shell regex could
eat a painted diagram once a later mermaid block appeared. Cache every
diagram in the HTML so v-html cannot flash it back to source.
Refs: https://github.com/Tencent/WeKnora/issues/2885
CI's system Python has packaging, so a false environment marker is skipped
silently instead of reported as a note. The install must still succeed; only
the note is conditional on whether markers can be evaluated.
The gate was answering the wrong question (every .py as isolated __main__),
and the installer derived dependencies from SKILL.md prose independently.
Align import prefixes with Python's path rules, keep the gate as the only
authority, and feed its findings back for one repair round.
Unset (0) now resolves at call time (2048 / 4096 / 24576 with sandbox) so the editor can offer Default vs Custom without silently rewriting a saved value.
ClawHub's download API keys by the skill slug alone. Sending
owner/slug as the slug query 404'd valid listings such as
@jixinyi546-maker/emar-ppt-skill.
Generated PPT/report scripts overflowed the shell_exec command cap, and
session agents kept mutating the frozen skill tree or failing SkillHub
installs on display titles. Add write/edit sandbox tools, attach the
skill venv to /workspace scripts, and accept slug-based install names.
list/read sandbox tools were advertised for skills-off agents so they could inspect /workspace/input, but they still refused anything outside /workspace/output. Register shell_exec independently of the skills manager so a skills-enabled sandbox with no ready skill gets a shell without empty read_skill / execute_skill_script tools.
shell_exec was gated on "skills enabled AND a ready skill exists", so an
agent whose skills were still installing (or had none) never received a
shell. Since shell_exec can execute skill scripts, it now follows the
SkillsEnabled switch alone.
list_sandbox_files / read_sandbox_file are pure filesystem capabilities
and no longer follow SkillsEnabled at all: they are registered whenever
the sandbox advertises a SessionFileStore, including for the installer.
MessageAttachment.URL is json:"-" so it never survives DB serialization,
but stageSessionAttachments depended on it to read the source file, which
silently dropped every attachment before it could be written into
/workspace/input.
Resolve the storage handle from the temporary_documents row keyed by
attachment.ID (GetScoped) when the URL is missing, and pass the session
tenant through staging so shared-agent sessions look up the right row.
* fix(wiki): prevent concurrent duplicate page identities
* fix(wiki): keep identity claims stable across LLM merges and Redis
Only exact same-title pages may overwrite a Redis identity claim. Successful
Redis reads now refresh the batch-local map so concurrent batches stay
aligned, citation slugs are reclaimed, and Reduce remaps onto the claimed
slug before locking.
* fix(wiki): batch identity lookups and harden claim convergence
Exact title resolution was one seq-scan per extracted item, and a dirty Redis claim could split concurrent workers. Batch the lookup, memoize it per batch, overwrite invalid claims, and keep remapped Item.Slug in sync.
---------
Co-authored-by: wizardchen <wizardchen@tencent.com>
Skill steps were hard to tell apart in the stream, and list_sandbox_files fell through to raw output. Quote the skill or path in the title and add dedicated cards for read_skill and sandbox listings.
Stop treating a failed sandbox as a full success, avoid writing a SHA
when the catalog archive is missing, and pick the newest stored fork
on migrate so files and chips match the catalog users actually see.
A skill is a workspace definition, not a sandbox setting. Keep one catalog
row and run a full image install for each backend, so agents only enable
skills that are already ready on the sandbox they use.
Failed execute_skill_script results were reduced to "exited with code 1" plus a retry hint, so the chat card hid the real stderr/stdout and the next model round could not correct arguments like a missing chart axis.
Installing skills was buried as a wizard step. Give it a first-class settings page so admins can browse and install without opening the sandbox drawer, and keep rollout on the runtime step.
Only show the key when the skill declared env vars, and open the editor
as a popup like the install transcript. Hide the hover tooltip while the
popup is open and keep password managers from filling the fields.
The installer reserved the entire WEKNORA_ prefix, so names like
WEKNORA_API_KEY were dropped while install still succeeded. Reserve only
the sandbox-injected names (and WEKNORA_SKILL_*), and keep that list in
one place.
ContainerStart returning success does not mean State.Running, so the first
skill-install ExecCreate can 409. Wait for running after start, and retry
once if exec still races.
* feat(chat): add a question outline for long overflowing sessions
When the thread overflows, show a centered tick rail of loaded user questions on the right. Hover opens the list; click jumps to that message. Pure frontend, no API change.
* feat(chat): move question outline left with mountain hover and Q&A preview
Put the tick rail on the left inner edge. Hover grows nearby ticks like a mountain and shows a following card with that turn's question plus an answer preview.
* fix(chat): allow offsetTop on minimap tick mapping input
The type-check CI failed because tests pass measured items with offsetTop into mapQuestionTicks, which only accepted { id }.
Co-authored-by: Cursor <cursoragent@cursor.com>
---------
Co-authored-by: hailongzhao <hailongzhao@tencent.com>
Co-authored-by: Cursor <cursoragent@cursor.com>
The endpoints this PR added were reachable but undescribed: Swagger had no
/me/env-vars paths, docs/api/skill.md never mentioned the admin PATCH, and
the Go SDK could neither set a workspace value nor a personal one, so an
integration had to hand-roll the calls.
Regenerates the swag artifacts and adds the missing client methods, plus
the caveat that matters most in practice: values are stored per calling
identity, so a personal value entered in the web UI never applies to a run
driven by an API key.
Capture existed so a credential a user typed in chat survives the turn.
It also overwrote values already stored, inferred the skill from a path in
the command, and logged the command that carried the secret — so one
hallucinated `export KEY=test` could permanently replace a working key,
and the key ended up in the application log either way.
Capture now only fills a blank: a name the workspace or the caller has
already filled in is dropped before the write path sees it, the skill must
be named explicitly, and assignment values are masked in logs.
The path the error message advertised did not exist: a required variable
was refused before the call's own env was considered, so an IM user, who
has no settings page, could never get the skill to run. Values supplied by
the call now satisfy the requirement, and the message names the tool that
accepts them.
Also take envs out of the generic UpdateSkill. Every install-progress
write is a read-modify-write of the whole row, and the heartbeat reloads
it every 30 seconds, so carrying envs along could put a stale declaration
back over one just recorded. Encryption failures abort the write instead
of storing plaintext in a queryable JSONB column, matching TenantAPIKey.
Reverts an unrelated deletion of the no-volume-support warning.
Skills declare the environment variables they read at run time, so a
credential no longer has to be an admin-set value shared by everyone
using the same skill image.
The installer agent writes a declaration the server validates against the
bundle. Admins fill workspace-wide values from the skill drawer, members
fill their own under Settings, and a successful shell_exec captures the
declared names a run already used for the current speaker. Resolution is
keyed by Principal so IM callers do not collapse onto one synthetic
account, and values are injected only into the process of that one run.
Values are AES-GCM encrypted at rest and never returned by any endpoint.
ArtifactCollector only drains /workspace/output from the pinned sandbox
config. Docker is session-persistent like Cube/E2B, but execution skipped
the pin, so generated HTML showed as unavailable after the turn.
Dedicated rerank already scores the recall set; falling back to the
conversation model mixed RRF scores with [0,1] ratings and could
resurface documents the reranker had rejected.
* feat(agent): preview skill artifacts and verify installs without running them
Generated files open in the chat drawer with overlay/Esc returning to the
list, and install verification no longer executes guessed skill entry
points. A failed install can be retried from the saved archive.
* ui(chat): use a folder icon and corner count for skill artifacts
The eye and document glyphs read as preview or copy, and TDesign's
loading spinner sat beside the icon. A folder with a top-right count
matches the generated-files drawer without covering the toolbar action.
* feat(artifact): implement inline artifact references with previews
Add support for inline artifact references in Markdown, allowing generated files to be displayed as clickable cards. Implemented a new rendering mechanism for artifacts, enabling previews directly from the chat interface. Updated localization files for new inline preview hints and missing file messages. Introduced tests for artifact reference normalization and rendering logic.
* fix(client): document AgentResponseType constants for revive
Touching the const block made golangci-lint require comments on every
exported value. Match the rest of the SDK and describe artifacts_pending.
* fix(agent): wrap artifact-reference prompt lines for lll
Pre-push golangci-lint rejected the branch on three lines over 120
characters in the sandbox: reference guidance.
* fix(agent): scope HTML scripts and tighten install verification
Keep scripted HTML preview on skill artifacts only, recreate sandbox
workspace dirs after they are deleted, and treat nested skill files as
dependencies rather than first-party imports.
* refactor(artifact): enhance artifact reference handling and metadata
Updated artifact metadata to include a stable resource handle for better identification. Adjusted artifact reference resolution to support both handle and name forms, ensuring compatibility with existing references. Enhanced tests to validate the new handling logic and ensure proper rendering of artifacts in chat messages. Improved documentation for clarity on artifact reference formats and their usage in the system.
shell_exec with display_type was collapsed to a one-line omit placeholder,
so the next turn and a history reload lost stdout/stderr. Keep a capped
copy of the streams in both Output and Data.