Closes#3192
Root cause
----------
The artifact preview header is driven by a Radix Select whose <SelectValue>
renders the label of the <SelectItem> matching the current value. The option
list was built solely from `artifacts` in ArtifactsContext, which is only
synced from `thread.values.artifacts` (chat-box.tsx). Artifacts surfaced via
the message-layer `present_files` tool call are never written back to
`thread.values.artifacts`, so when such a file is selected its filepath has
no matching <SelectItem>. Radix then renders an empty trigger and the header
filename appears blank, even though the preview body loads correctly.
Fix
---
Compute `artifactOptions` as a defensive union: if the currently selected
filepath is missing from `artifacts`, prepend it so a matching <SelectItem>
always exists. This keeps the header label in sync with the active file
without changing context semantics or coupling the UI to a future
auto-discovery mechanism (see existing TODO in chat-box.tsx).
Tests
-----
Add an e2e case that mocks a thread whose only artifact is delivered via
`present_files` (thread.values.artifacts = []) and asserts both the header
title and preview content render. All 21 e2e tests pass.
* fix(agents): sync agent_name across context/configurable and reject empty soul (#3549)
Two independent issues caused custom agent creation to silently fail:
1. build_run_config only wrote agent_name into one container (configurable
or context), so setup_agent — which reads ToolRuntime.context exclusively
since LangGraph >=1.1.9 — saw agent_name=None and wrote SOUL.md to the
global base_dir instead of users/{user_id}/agents/{name}/. Mirror the
dual-write pattern already used by merge_run_context_overrides and
naming.py so both containers always carry the same value.
2. setup_agent persisted whatever soul string it received, including empty
or whitespace-only content, and still reported success. The frontend
then surfaced an unusable agent and the global default SOUL.md could be
silently overwritten with empty content. Reject empty soul before any
filesystem operation so the model can retry.
Tests:
- test_gateway_services.py: dual-write regressions for both configurable
and context entry paths, explicit-agent-name precedence on both sides,
and a shape-parity test against merge_run_context_overrides.
- test_setup_agent_tool.py: empty/whitespace soul rejection, plus
no-overwrite guarantees for existing global and per-agent SOUL.md.
* Update services.py
GET /api/mcp/config returns 403 for non-admin users, but the previous
client returned the error body as MCPConfig, causing MCPServerList to
crash with 'Cannot convert undefined or null to object' on
Object.entries(config.mcp_servers).
- api.ts: introduce MCPConfigRequestError; loadMCPConfig and
updateMCPConfig now throw it (carrying status + isAdminRequired)
instead of letting non-2xx bodies leak through as parsed config
- tool-settings-page.tsx: render a friendly 'admin privileges required'
empty state when the React Query error is an admin-required
MCPConfigRequestError; keep MCPServerList resilient with
Object.entries(servers ?? {}) and an empty-state for no servers
- i18n: add settings.tools.adminRequired and settings.tools.empty in
en-US, zh-CN and the Translations type
- tests: cover 403 / 5xx / instanceof / detail-fallback for both
loadMCPConfig and updateMCPConfig in tests/unit/core/mcp/api.test.ts
Refs: #3527
* fix(frontend): keep workspace interactive when SSR auth probe cannot reach gateway (#3493)
When the SSR auth probe at /api/v1/auth/me times out or fails, the
workspace layout used to render a static fallback page without
AuthProvider or QueryClientProvider, making logout and every other
interaction non-functional until the gateway recovered.
Render the normal WorkspaceContent in 'gateway_unavailable' mode
instead, surfacing a polite offline banner that re-probes the gateway
in the background and hides itself the moment refreshUser() returns
an authenticated user. The probe is reentrancy-guarded so a slow
gateway cannot pile up parallel /auth/me requests.
Closes#3493
* fix(workspace): silent probe in offline banner to avoid /login redirect during gateway recovery (#3493)
The banner previously delegated retry probes to AuthProvider.refreshUser(),
which treats any 401 from /api/v1/auth/me as 'session expired' and
force-redirects to /login. During gateway recovery, the first few requests
may transiently return 401 before the gateway is fully ready, which would
incorrectly kick the user out — defeating the purpose of the offline banner.
Now the banner silently fetches /api/v1/auth/me itself and only delegates
to refreshUser() on 200 OK. Non-200 responses (401 / 5xx / network) are
swallowed and retried on the next interval tick, ensuring the user stays
logged in across short gateway outages.
Verified in Docker:
- docker pause deer-flow-gateway → banner appears, page interactive
- docker unpause deer-flow-gateway → banner auto-disappears within 10s,
user remains on /workspace/chats/new with full session restored
- All 117 unit tests pass
* fix(workspace): fix banner polling leak and persistent 401 handling (#3493)
- Stop polling immediately after user recovery: add user to effect dependencies, cleanup interval when user !== null
- Handle persistent 401: trigger login redirect after 3 consecutive unauthorized responses
- Extract decision logic to pure helper, add 8 unit tests covering all critical paths
* fix(workspace): address CR feedback on gateway offline recovery (#3493)
- gateway-offline-banner-helpers: decrement (not reset) auth-failure
streak on transient outcomes so a flapping gateway (401 alternating
with 5xx) still converges on session-expired
- gateway-offline-banner: reuse probe response body to apply user
directly via new AuthProvider.applyUser, halving the recovery burst
against an already-struggling gateway
- gateway-offline-banner: extract classifyProbe into helpers for unit
testability; log probe failures via console.warn instead of swallowing
- gateway-offline-fallback: new shared component used by both workspace
and (auth) layouts so auth pages recover the same way the workspace
does, fixing the lockup where bare static HTML had no AuthProvider
- AuthProvider.logout: fall back to hard navigation when the gateway
logout fetch fails, matching legacy form-POST behaviour and avoiding
stale client state during outage
- tests: extend gateway-offline-banner-helpers.test with flapping
convergence and classifyProbe branch coverage (19 cases total)
* fix(sandbox): persist lazily-acquired sandbox state via Command
ensure_sandbox_initialized mutates runtime.state in place, which is local
to the current tool invocation and is not picked up by LangGraph's channel
reducer. Subsequent graph steps and downstream consumers (such as
ToolOutputBudgetMiddleware and the sub-agent task_tool) therefore cannot
observe the sandbox id from state.
Wrap tool calls in SandboxMiddleware (wrap_tool_call / awrap_tool_call) to
detect fresh lazy initialization by diffing runtime.state before and after
the handler, and emit a proper state update via Command(update=...):
- ToolMessage results are wrapped into Command(update={sandbox, messages})
- Command results with a dict update are merged on the sandbox key while
preserving messages / goto / graph / resume
- Command results with non-dict updates are left untouched to avoid silent
data loss on unknown update shapes
Tests:
- 7 new unit tests cover lazy-init emit, passthrough, dict-update merge,
non-dict-update passthrough (sync and async)
- Refresh replay golden write_read_file.ultra.events.json: SSE 'values'
events now correctly carry the 'sandbox' key in their keys list, which
is the direct evidence that the fix is effective
Closes#3463
* refactor(sandbox): use dataclasses.replace to preserve Command fields
Address Copilot review on #3464: replace manual field-copy with
dataclasses.replace so any current or future Command fields are
preserved automatically when merging sandbox_update.
Also add a regression test that constructs a Command with non-None
graph/goto/resume to lock this behavior in.
* fix(frontend): paginate workspace chat list beyond 50 threads (#3482)
The sidebar 'Recent chats' and /workspace/chats list were hard-capped
at the first 50 threads returned by threads.search. Replace the
single-shot useThreads() consumers with useInfiniteThreads() and add
an IntersectionObserver sentinel to each list so further pages are
fetched on demand.
In search mode on the chats page, the sentinel is replaced by an
explicit 'Load more' button to prevent the observer from draining the
entire backend list while the filtered view stays empty.
- Add useInfiniteThreads + page-size constant and pure cache helpers
(map/filterInfiniteThreadsCache, getInfiniteThreadsNextPageParam)
- Mirror rename / delete / stream-finish updates into the new
infinite cache so optimistic UI stays consistent
- Extend the e2e mock to honour limit/offset slicing
- Unit tests for the cache helpers and pagination boundary
- Playwright e2e covering chats page + sidebar load-more, and the
search-mode guard against runaway auto-pagination
- Add en/zh i18n entries for the search-mode load-more button
Fixes#3482
* docs(frontend): clarify infinite-threads offset semantics and test post-delete invariant
- Add docstring to getInfiniteThreadsNextPageParam explaining that TanStack
Query freezes the returned offset into pageParams once, so optimistic cache
mutations that shrink page lengths (filterInfiniteThreadsCache on delete)
cannot retroactively move the offset backwards. Delete/rename paths
reconcile against the backend via invalidateQueries in onSettled.
- Add unit test covering the post-delete invariant.
- Fix misleading comment in thread-list-infinite-scroll.spec.ts: the
thread-search mock does not sort by updated_at; it returns the array in
the order provided.
Addresses Copilot CR comments on #3485.
* fix(frontend): mirror onCreated upsert into infinite cache; add sidebar Load-older button
Address review feedback on #3485:
- New upsertThreadInInfiniteCache helper; useThreadStream onCreated now
upserts into both the legacy ['threads','search'] cache and the new
infinite cache, so a freshly created thread appears in the sidebar
immediately during streaming instead of only after the run finishes
and onSettled invalidates the query. Restores parity with main.
- Sidebar Recent Chats now exposes a visible 'Load older chats' button
alongside the IntersectionObserver sentinel, so keyboard-only users
and environments where IO is unavailable can still reach older
conversations.
- Add zh-CN / en-US / types entry for chats.loadOlderChats.
- Cover the new helper with 3 unit tests (no-op on uninitialised cache,
prepend new thread to first page, merge with existing entry without
duplication).
* fix(middleware): externalize oversized tool output into sandbox for non-mounted sandboxes
ToolOutputBudgetMiddleware persisted oversized tool results to the host
filesystem and returned a /mnt/user-data/outputs virtual path. For sandboxes
that do not use thread-data mounts (e.g. remote AIO sandbox), that virtual
path does not exist inside the sandbox, so the model's read_file tool could
not read it back and reported 'file not found'.
Branch on SandboxProvider.uses_thread_data_mounts:
- Mounted sandboxes (local Docker, AIO + LocalContainerBackend) keep the
original host-disk path; the host outputs dir is bind-mounted to the same
virtual path inside the sandbox, so behavior is unchanged.
- Non-mounted (remote) sandboxes externalize into the sandbox itself via
execute_command('mkdir -p ...') + write_file + 'test -s' validation. The
validation step is required because AIO sandbox execute_command returns
'Error: ...' as a string on failure instead of raising, so a silent mkdir
failure would otherwise leak through.
Any failure (rejected subdir, mkdir/write/validate error) falls back to the
existing inline head+tail truncation, so an unreadable path is never returned
to the model.
The sandbox resolver reads the sandbox_id that SandboxMiddleware already
writes into runtime.state['sandbox']; it never calls provider.acquire(),
keeping the tool-call hot path free of blocking I/O. Tools that do not use a
sandbox (web_search, MCP, ...) resolve to None and fall through to inline
truncation, which is the safe behavior for them.
Fixes#3416
* fix(middleware): address Copilot review feedback on sandbox externalization
- Make get_sandbox_provider() lookup best-effort in _budget_content: only
query when outputs_path or sandbox is available, and fall back to inline
truncation if provider initialization raises rather than propagating
the error. A resolved sandbox instance is sufficient on its own to take
the non-mounted externalization branch.
- Strict-match the sandbox post-write validation echo
(check.strip() == 'OK') to avoid false positives if execute_command
ever surfaces unrelated stdout/stderr containing 'OK' as a substring.
Refs: #3417
* test: fix flaky tests relying on /nonexistent/... path under container root
Two tests in this module (test_returns_none_on_invalid_path and
test_fallback_when_disk_write_fails) used paths like
'/nonexistent/impossible/path' to trigger _externalize's OSError
fallback. These paths are creatable when the test process runs as root
inside the CI container: os.makedirs(..., exist_ok=True) successfully
creates the entire chain under /, so the OSError branch is never hit
and the tests fail. Reproducible on main independently of this PR.
Switch to '/dev/null/cannot-mkdir-here'. /dev/null is a character
device on both Linux and macOS, so os.makedirs always fails with
NotADirectoryError regardless of privileges, reliably exercising the
OSError fallback.
* fix(tool-output-budget): only consult sandbox provider when a sandbox is resolved
The previous revision called get_sandbox_provider() whenever externalization
was triggered, including on the legacy host-disk path. Environments without
a configured sandbox -- in particular CI runners without a config.yaml --
would raise FileNotFoundError there, get caught, and silently fall back to
inline truncation. That defeated the host-disk externalization path that
predates this PR and was the root cause of the regressing legacy tests.
Restructure the branching so the provider is only consulted when a sandbox
has actually been resolved for the current tool call:
- sandbox resolved + provider.uses_thread_data_mounts: host-disk write
(bind-mounted into the sandbox, equivalent to a sandbox-side write).
- sandbox resolved + non-mounted provider: sandbox write (#3416).
- no sandbox + outputs_path: host-disk write
(legacy / non-sandbox tools, no provider call at all).
- otherwise: inline fallback.
No test changes; the legacy externalization tests are provider-agnostic by
construction and now pass without monkeypatching.
Refs: #3416
* test(tool-output-budget): assert legacy path does not call sandbox provider
Lock in the contract introduced by d6e2d25b: when no sandbox is resolved
for a tool call, _budget_content must externalize to the host outputs
directory without consulting get_sandbox_provider(). Regressing this would
re-break legacy / non-sandbox tools in environments without a configured
sandbox (e.g. CI without config.yaml), which is the failure mode #3416's
fix avoids.
The test injects a get_sandbox_provider that raises on call, so any
future refactor that moves the provider lookup out of the sandbox-only
branch will fail loudly.
Refs: #3416
* fix(#3189): prevent write_file streaming timeout on long reports
Adds a layered defense against StreamChunkTimeoutError caused by oversized
single-shot write_file tool calls:
- factory: default stream_chunk_timeout to 240s for OpenAI-compatible
clients (overridable via ModelConfig.stream_chunk_timeout in config.yaml)
- sandbox/tools: server-side 80 KB length guard on non-append write_file
calls (configurable via DEERFLOW_WRITE_FILE_MAX_BYTES env var, 0 disables);
rejects oversized payloads with a structured error pointing the model at
str_replace or append=True
- middleware: classify StreamChunkTimeoutError as transient but cap retries
at 1 via per-exception _RETRY_BUDGET_OVERRIDES (same-payload retry on a
chunk-gap timeout buffers the same way upstream; full 3-attempt loop
would stack 6-12 min of dead air)
- middleware: surface an actionable user-facing message for stream-drop
exceptions instead of leaking the raw langchain stack
- prompts: add a routing-style File Editing Workflow hint to both lead_agent
and general_purpose subagent prompts, pointing the model at str_replace
for incremental edits (mirrors Claude Code's Edit / Codex's apply_patch)
- tests: behavioural coverage for size guard, retry budget override,
stream-drop user message, factory default injection
Refs #3189
* fix(#3189): drop stream_chunk_timeout for non-OpenAI providers
Address CR feedback on PR #3195:
- factory: pop `stream_chunk_timeout` from kwargs for any model_use_path other than `langchain_openai:ChatOpenAI` instead of returning early. `ModelConfig.stream_chunk_timeout` is part of the shared schema, so a user-supplied value on a non-OpenAI provider would otherwise be forwarded to its constructor and raise `TypeError: unexpected keyword argument`.
- factory: rewrite docstring to describe the actual `exclude_none=True` behaviour (explicit null is excluded and falls back to the default) instead of the misleading "None falling out via exclude_none=True keeps its value".
- tests: add regression coverage asserting the kwarg is stripped before reaching a non-OpenAI provider's constructor.
Refs: bytedance#3189
* fix(#3189): restrict stream-drop user copy to StreamChunkTimeoutError only
Per CR on #3195: narrow _STREAM_DROP_EXCEPTIONS to StreamChunkTimeoutError. Generic httpx RemoteProtocolError / ReadError fall back to the standard 'temporarily unavailable' copy, since they routinely fire on transient network blips where the 'split the output' guidance is misleading. Retry/backoff classification is unchanged — both remain transient/retriable. Tests updated to reflect new copy, plus a symmetric regression test for ReadError.
---------
Co-authored-by: Willem Jiang <willem.jiang@gmail.com>
* fix(skills): surface offending line and quoting hint on SKILL.md YAML errors
When a SKILL.md front-matter fails to parse, the existing log only
echoes PyYAML's raw message, leaving authors to grep the file for the
offending line. This is especially painful for the very common
LLM-authored mistake of an unquoted scalar containing ': '
(e.g. 'description: foo: bar'), which fails with
'mapping values are not allowed here' and silently drops the skill.
Enrich the error log with:
- the source line PyYAML pointed at via problem_mark
- a targeted, copy-pasteable quoting hint when (and only when) the
error is the well-known 'mapping values are not allowed' scanner
error on an unquoted value
The skill is still rejected (no semantics are guessed or rewritten);
only the diagnostic is improved.
Fixes#3333
* improve(skills): address CR feedback on SKILL.md YAML error diagnostics
Per review on #3335:
- Log the file line number (mark.line + 2) instead of the
front-matter-internal line number, so authors land on the right
row in their editor.
- Use exc.problem == "mapping values are not allowed here" for a
tighter match than substring-scanning str(exc).
- Preserve the offending key's leading whitespace in the quoting
hint so nested mappings stay nested when authors paste the fix
back.
- Rewrite the regression test to actually exercise the new
behaviour: PyYAML's own message already echoes the offending
line (and truncates it with "..."), so the old assertion
passed on main. New assertions pin (a) the file-line number,
(b) the full untruncated line, and (c) the copy-pasteable hint.
- Add a guard test for nested-key indentation so the
partition()/strip() shape cannot regress silently.
Refs #3333, #3335
* fix(skills): escape backslashes in YAML quoting hint
The hint emitted by _format_yaml_error previously escaped only double
quotes, so values containing backslashes (e.g. Windows paths like
C:\Temp or regex escapes like \d) produced a suggested scalar that
was either invalid YAML or silently re-interpreted by PyYAML's
double-quoted escape rules when pasted back. Escape order matters:
backslashes first, then double quotes.
Adds two regression tests covering Windows-path and regex-style
backslashes.
Address Copilot CR feedback on PR #3335.
* fix(frontend): preserve chronological order of thread history after context compression
Iterate runs from newest to match backend `list_by_thread` (newest-first) and the prepend semantics of the history loader, so refreshed history renders in A→B→C→D→E→F order.
Fixes#3352
* fix(frontend): auto-continue loading runs with no visible messages after context compression
* fix(agents): preserve todos state across node updates
ThreadState.todos had no reducer, so any downstream node returning a
partial state without todos was implicitly setting it to None, which
LangGraph then used to overwrite the previously streamed value. This
caused the to-do list to render correctly during streaming but vanish
once streaming completed.
Add a merge_todos reducer that keeps the last non-None value, mirroring
the merge_artifacts pattern already used in the same file. An explicit
empty list is still respected so that 'user cleared todos' works.
Tests: 10 new unit tests in tests/test_thread_state_reducers.py covering
merge_todos plus regression coverage for merge_artifacts and
merge_viewed_images. All 69 thread-related tests pass locally.
Closes#3123
* test(agents): add annotation binding regression guard
Address Copilot review feedback on #3123:
- Add TestThreadStateAnnotations asserting that ThreadState.todos is
Annotated with merge_todos. Without this guard, reverting the
Annotated[list | None, merge_todos] binding would silently regress
#3123 while all existing reducer unit tests continue to pass.
- Align test imports to 'from deerflow.agents.thread_state import ...'
matching the rest of the backend test suite.
