`get()` accepts `workspace_id: str | None | _AutoSentinel = AUTO` and
moves the workspace check into the SQL WHERE so cross-workspace lookups
short-circuit without loading the row. `MemoryThreadMetaStore._get_owned_record`
gets the same filter for parity.
The user_id check stays as a post-load comparison (preserves the existing
shared-row semantics where row.user_id IS NULL means "everyone in this
workspace"). Cross-workspace always returns None, never the row.
3 new tests cover the three states: in-workspace get returns the row,
out-of-workspace get returns None even when user_id matches, explicit
workspace_id=None bypasses (migration / CLI). Existing 22 thread_meta
tests stay green.
`create()` now accepts `workspace_id: str | None | _AutoSentinel = AUTO`
on both the SQL and in-memory implementations (and the abstract base).
AUTO resolves via `resolve_workspace_id()` from the workspace contextvar
that PR4 AuthMiddleware sets; explicit None bypasses for migration paths;
explicit str overrides the contextvar.
Test infrastructure:
- conftest gains an autouse `_auto_workspace_context` fixture mirroring
the existing user fixture. Opt-out via `@pytest.mark.no_auto_workspace`.
- A SQLAlchemy `after_create` listener on `Base.metadata` seeds the
matching `test-workspace-autouse` + `test-user-autouse` rows whenever
`init_engine` runs `create_all()`, so the FK from threads_meta to
workspaces resolves. Alembic migration tests bypass create_all and are
unaffected, keeping real FK constraints under test.
- `test_thread_meta_workspace_filter.py` covers the three AUTO / explicit
/ None paths.
3 new tests pass; 115 existing thread_meta/run/feedback/run_event/owner
tests stay green.
Refuses to upgrade if any of the 4 business tables still has NULL
workspace_id rows (pre-flight count then RuntimeError pointing the
operator at scripts/backfill_workspace_id.py). Then op.batch_alter_table
flips each column to NOT NULL and adds the threads_meta
(workspace_id, thread_id) UNIQUE index that PR6 routers will rely on.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
ORM-side mirror of alembic 0002. threads_meta gets the matching
(workspace_id, user_id, updated_at) composite index so fresh
create_all() dev databases land on the same shape Alembic produces.
Pre-existing test_alembic_default_workspace_id pins to revision
0001 explicitly since 'head' now includes 0002.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
ALTER threads_meta / runs / feedback / run_events to add nullable
workspace_id String(36) + FK to workspaces (ON DELETE CASCADE) and a
composite (workspace_id, user_id, updated_at) index on threads_meta for
the workspace-scoped thread-list query. Downgrade reverses all four FKs,
columns, and the index.
Stage 0 PR5 step 1/2; revision 0003 flips NOT NULL after the backfill.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
initialize_admin now seeds a 1-person workspace immediately after the
admin user is created: WorkspaceRepository.create(name, slug, owner_id)
+ WorkspaceMembershipRepository.add(role='owner') + writes the new
workspace id back to users.default_workspace_id. The session JWT is
re-issued with wid + role='owner' so subsequent requests pass the
T4.7 workspace gate.
Mechanical pieces:
- SQLiteUserRepository row<->user mapping now includes
default_workspace_id (sql update_user too) so the column persists.
- workspace.sql.SLUG_BLACKLIST is now public (was _SLUG_BLACKLIST) and
the registration helper treats blacklisted slugs as "taken" so the
walker steps past reserved names like "admin" instead of crashing.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Adds a nullable FK column to mirror alembic 0001. On fresh deployments
metadata.create_all() will create users with this column; existing
deployments rely on the alembic migration to add it.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
First Alembic revision for the DeerFlow application schema. Adds a
nullable users.default_workspace_id column with an ON DELETE SET NULL
FK to workspaces(id), so newly-registered users can be routed back to
their default workspace on next login without consulting memberships.
Uses op.batch_alter_table for SQLite ALTER compatibility (sqlite is
still a supported dev fallback).
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
* fix(task): remove max_turns parameter from task tool interface
Subagents should always use their configured max_turns value. Exposing
this parameter allowed callers to override the admin-configured limit,
which is undesirable. The value is now exclusively driven by subagent
config (per-agent overrides and global defaults in config.yaml).
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
* Potential fix for pull request finding
Co-authored-by: Copilot Autofix powered by AI <175728472+Copilot@users.noreply.github.com>
---------
Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
Co-authored-by: Willem Jiang <willem.jiang@gmail.com>
Co-authored-by: Copilot Autofix powered by AI <175728472+Copilot@users.noreply.github.com>
* fix(tools): introduce Runtime type alias to eliminate Pydantic serialization warning
Add deerflow/tools/types.py with:
Runtime = ToolRuntime[dict[str, Any], ThreadState]
Replace every runtime: ToolRuntime[ContextT, ThreadState] and
runtime: ToolRuntime[dict[str, Any], ThreadState] annotation in
sandbox/tools.py, present_file_tool.py, task_tool.py, view_image_tool.py,
and skill_manage_tool.py with the new Runtime alias.
The unbound ContextT TypeVar (default None) caused
PydanticSerializationUnexpectedValue warnings on every tool call because
LangChain's BaseTool._parse_input calls model_dump() on the auto-generated
args_schema while DeerFlow passes a dict as runtime context.
Binding the context to dict[str, Any] aligns Pydantic's serialization
expectations with reality and removes the noise from all run modes.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Co-authored-by: Cursor <cursoragent@cursor.com>
* fix(tools): extend Runtime alias to setup_agent and update_agent tools
Replace bare ToolRuntime annotations in setup_agent_tool.py and
update_agent_tool.py with the shared Runtime alias introduced in the
previous commit, and add both tools to the Pydantic serialization
warning regression test (13 cases total).
Co-authored-by: Cursor <cursoragent@cursor.com>
* test(tools): loosen Pydantic warning filter to avoid version-specific format
Replace the brittle "field_name='context'" substring check with a looser
"context" match so the assertion stays valid if Pydantic changes its
internal warning format across versions.
Co-authored-by: Cursor <cursoragent@cursor.com>
* test(tools): simplify warning filter and clean up docstring
Remove the "context" substring condition from the Pydantic warning
filter — asserting that no PydanticSerializationUnexpectedValue fires
at all is both simpler and more comprehensive, since the test payload
contains only the tool's own args plus runtime.
Also update the module docstring to remove the version-specific warning
format example that was inconsistent with the looser filter.
Co-authored-by: Cursor <cursoragent@cursor.com>
---------
Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
Co-authored-by: Cursor <cursoragent@cursor.com>
* Make loop detection configurable
Expose LoopDetectionMiddleware thresholds through config.yaml while preserving existing defaults and allowing the middleware to be disabled.
Refs bytedance/deer-flow#2517
* feat(loop-detection): add per-tool tool_freq_overrides to Phase 1
Adds ToolFreqOverride model and tool_freq_overrides field to
LoopDetectionConfig, wires it through LoopDetectionMiddleware, and
documents the option in config.example.yaml.
Resolves the gap flagged in the #2586 review: without per-tool overrides,
users hit by #2510/#2511 (RNA-seq workflows exceeding the bash hard limit)
had no way to raise thresholds for one tool without loosening the global
limit for every tool.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
* Potential fix for pull request finding
Co-authored-by: Copilot Autofix powered by AI <175728472+Copilot@users.noreply.github.com>
* docs(loop-detection): document tool_freq_overrides in LoopDetectionMiddleware docstring
Add the missing Args entry for tool_freq_overrides, explaining the
(warn, hard_limit) tuple structure and how per-tool thresholds supersede
the global tool_freq_warn / tool_freq_hard_limit for named tools.
Also run ruff format on the three files flagged by the lint check.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
* fix(loop-detection): validate LoopDetectionMiddleware __init__ params eagerly
Raise clear ValueError at construction time instead of crashing at
unpack-time inside _track_and_check when bad values are passed:
- tool_freq_overrides: must be 2-tuples of positive ints with hard_limit >= warn
- scalar thresholds: warn_threshold, hard_limit, tool_freq_warn,
tool_freq_hard_limit must be >= 1 and hard limits must >= their warn pairs
- window_size, max_tracked_threads must be >= 1
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
* fix(test): isolate credential loader directory-path test from real ~/.claude
The test didn't monkeypatch HOME, so on any machine with real Claude Code
credentials at ~/.claude/.credentials.json the function fell through to
those credentials and the assertion failed. Adding HOME redirect ensures
the default credential path doesn't exist during the test.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
* style(test): add blank lines after import pytest in TestInitValidation
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
* refactor(loop-detection): collapse dual validation to LoopDetectionConfig
Modifications
- LoopDetectionMiddleware.__init__: stripped of all ValueError raises;
becomes a plain field-assignment constructor.
- LoopDetectionMiddleware.from_config: classmethod that builds the
middleware from a Pydantic-validated LoopDetectionConfig and handles
the ToolFreqOverride -> tuple[int, int] conversion.
- agents/factory.py: SDK construction routed through
LoopDetectionMiddleware.from_config(LoopDetectionConfig()) so the
defaults path is Pydantic-validated too.
- agents/lead_agent/agent.py: uses from_config instead of unpacking
config fields by hand.
- tests/test_loop_detection_middleware.py: deleted TestInitValidation
(16 methods exercising the removed __init__ checks); added
TestFromConfig (4 tests: scalar field mapping, override tuple
conversion, empty overrides, behavioral smoke test).
Result: one validation layer (Pydantic), zero duplication, no __new__
hacks. Both production construction sites flow through LoopDetectionConfig.
Test results
make test -> 2977 passed, 18 skipped, 0 failed (137s)
make format -> All checks passed; 411 files left unchanged
* feat(agents): make loop_detection configurable in create_deerflow_agent
Adds a `loop_detection: bool | AgentMiddleware = True` field to
RuntimeFeatures, mirroring the existing pattern used by `sandbox`,
`memory`, and `vision`. SDK users can now disable LoopDetectionMiddleware
or replace it with a custom instance built from their own
LoopDetectionConfig — e.g.
`LoopDetectionMiddleware.from_config(my_cfg)` — instead of being stuck
with the hardcoded defaults previously installed by the SDK factory.
The lead-agent path (which already reads AppConfig.loop_detection) is
unchanged, and the default `True` preserves prior always-on behavior for
all existing callers.
Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
---------
Co-authored-by: knight0940 <631532668@qq.com>
Co-authored-by: Claude Opus 4.7 <noreply@anthropic.com>
Co-authored-by: Amorend <142649913+knight0940@users.noreply.github.com>
Co-authored-by: Copilot Autofix powered by AI <175728472+Copilot@users.noreply.github.com>
Co-authored-by: Willem Jiang <willem.jiang@gmail.com>
* feat(agent): add update_agent tool for in-chat custom-agent self-updates (#2616)
Custom agents had no built-in way to persist updates to their own SOUL.md /
config.yaml from a normal chat — `setup_agent` was only bound during the
bootstrap flow, so when the user asked the agent to refine its description
or personality, the agent would shell out via bash/write_file and the edits
landed in a temporary sandbox/tool workspace instead of
`{base_dir}/agents/{agent_name}/`.
Changes:
- New `update_agent` builtin tool with partial-update semantics (only the
fields you pass are written) and atomic temp-file + os.replace writes so
a failed update never corrupts existing SOUL.md / config.yaml.
- Lead agent now binds `update_agent` in the non-bootstrap path whenever
`agent_name` is set in the runtime context. Default agent (no
agent_name) and bootstrap flow are unchanged.
- New `<self_update>` system-prompt section is injected for custom agents,
instructing them to use `update_agent` — and explicitly NOT bash /
write_file — to persist self-updates.
- Tests: 11 new cases in `tests/test_update_agent_tool.py` covering
validation (missing/invalid agent_name, unknown agent, no fields),
partial updates (soul-only, description-only, skills=[] vs omitted),
no-op detection, atomic-write safety, and AgentConfig round-tripping;
plus 2 new cases in `tests/test_lead_agent_prompt.py` covering the
self-update prompt section.
- Docs: updated backend/CLAUDE.md builtin tools list and tools.mdx
(en/zh) with the new tool description.
* feat(agent): isolate custom agents per user
Store custom agent definitions under the effective user, keep legacy agents readable until migration, and cover API/tool/migration behavior with tests.
Co-authored-by: Cursor <cursoragent@cursor.com>
* feat: consistent write/delete targets & add --user-id to migration
---------
Co-authored-by: Cursor <cursoragent@cursor.com>
* fix(loop-detection): keep tool-call pairing on warn injection (#2724)
* make format
* fix(loop-detection): avoid IMMessage leak to downstream consumer
* fix(channels): filter loop warning text from IM replies
* feat(community): add Serper web search provider
Add a new community search provider backed by the Serper Google Search
API (https://serper.dev). Serper returns real-time Google results via a
simple JSON API and requires only an API key — no extra Python package.
Changes:
- backend/packages/harness/deerflow/community/serper/__init__.py
- backend/packages/harness/deerflow/community/serper/tools.py
Implements web_search_tool using httpx (already a project dependency).
API key is read from config.yaml `api_key` field or SERPER_API_KEY env var.
Follows the same interface / output shape as the existing ddg_search provider.
Exposes max_results parameter (default 5) with config override logic.
- backend/tests/test_serper_tools.py
Unit tests covering API key resolution, config overrides, HTTP errors,
empty results, and parameter passing.
- config.example.yaml: add commented-out Serper example alongside other providers
- .env.example: add SERPER_API_KEY placeholder
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
* Fix the lint error
* Fix the lint error
---------
Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
Co-authored-by: Willem Jiang <willem.jiang@gmail.com>
* fix(gateway): return ISO 8601 timestamps from threads endpoints (#2594)
ThreadResponse documents created_at / updated_at as ISO timestamps,
matching the LangGraph Platform schema (langgraph_sdk.schema.Thread
exposes them as datetime, JSON-encoded as ISO 8601). The gateway
threads router was instead emitting str(time.time()) — unix-second
floats — breaking frontend new Date() parsing and producing a mixed
ISO/unix wire format that also corrupted the search sort order.
Centralize timestamp generation in deerflow.utils.time:
- now_iso() — datetime.now(UTC).isoformat()
- coerce_iso(x) — heals legacy unix-timestamp strings on read so the
store converges to ISO without a one-shot migration
threads.py: replace 6 time.time() call sites with now_iso(); wrap all
read paths and Phase-2 checkpoint metadata with coerce_iso(); _store_upsert
opportunistically heals legacy created_at on update; drop unused time import.
thread_runs.py: reuse now_iso() instead of a private duplicate _now_iso(),
preventing future drift between the two timestamp call sites.
Tests: 9 unit tests for the helper; 5 integration tests pinning the ISO
contract for create/get/patch/search and the legacy-healing path on the
internal store upsert. Full suite: 2144 passed, 15 skipped, 0 failed.
Closes#2594
* fix(gateway): coerce checkpoint metadata timestamps to ISO on read
After the merge with main, three additional read paths in ``threads.py``
were still emitting raw ``str(metadata.get("created_at", ""))`` —
``get_thread_state``, ``update_thread_state``, and ``get_thread_history``.
Same root cause as #2594: when the checkpoint metadata's ``created_at``
is a unix-second float (legacy data, or a checkpoint written by an older
Gateway version), ``str(float)`` produces ``"1777252410.411327"`` and the
frontend's ``new Date(...)`` returns ``Invalid Date``. The fix on the
``/threads/{id}`` GET path was already in place; these three sibling
endpoints needed the same treatment.
All four call sites now flow through ``coerce_iso``, so:
- legacy float metadata heals to ISO on the way out,
- ISO metadata passes through unchanged,
- ``datetime`` instances (which the new ``coerce_iso`` branch handles
explicitly) emit with the ``T`` separator instead of falling through
to the space-separated ``str(datetime)`` form.
Coverage added for the two endpoints not already pinned by the merge:
- ``test_get_thread_state_returns_iso_for_legacy_checkpoint_metadata``
- ``test_get_thread_history_returns_iso_for_legacy_checkpoint_metadata``
Both pre-seed a checkpoint whose metadata carries the literal float
from the issue body and assert the wire format is ISO.
* refactor: thread app config through lead prompt
* fix: honor explicit app config across runtime paths
* style: format subagent executor tests
* fix: thread resolved app config and guard subagents-only fallback
Address two PR review findings:
1. _create_summarization_middleware passed the original (possibly None)
app_config into create_chat_model, forcing the model factory back to
ambient get_app_config() and risking config drift between the
middleware's resolved view and the model's view. Pass the resolved
AppConfig instance through end-to-end.
2. get_available_subagent_names accepted Any-typed config and forwarded
it to is_host_bash_allowed, which reads ``.sandbox``. A
SubagentsAppConfig (also accepted upstream as a sum-type input) has
no ``.sandbox`` attribute and would be silently treated as "no
sandbox configured", incorrectly disabling the bash subagent. Guard
on hasattr and fall back to ambient lookup otherwise.
Adds regression tests for both paths.
* chore: simplify hasattr guard and tighten regression tests
- Collapse if/else into ternary in get_available_subagent_names; hasattr(None, ...) is False so the explicit None check was redundant.
- Drop comments that narrate the change rather than explain non-obvious WHY (test names already convey intent).
- Replace stringly-typed sentinel "no-arg" in regression test with direct args tuple comparison.
---------
Co-authored-by: greatmengqi <chenmengqi.0376@bytedance.com>
* fix(sandbox): pass no_change_timeout to exec_command to prevent 120s premature termination
The agent_sandbox library's shell API defaults no_change_timeout to 120
seconds. When AioSandbox.execute_command() called exec_command() without
this parameter, commands producing no output for 120s would return with
NO_CHANGE_TIMEOUT status even though the script was still running.
Pass no_change_timeout=600 to all exec_command calls (matching the
client-level HTTP timeout) so long-running commands are not cut short.
Fixes#2668
* test(sandbox): add assertions for no_change_timeout in execute_command and list_dir
Agent-Logs-Url: https://github.com/bytedance/deer-flow/sessions/2f37bc72-0826-4443-a6ba-e5b78c22fb5a
Co-authored-by: WillemJiang <219644+WillemJiang@users.noreply.github.com>
---------
Co-authored-by: copilot-swe-agent[bot] <198982749+Copilot@users.noreply.github.com>
* fix(subagents): use model override for tools and middleware
* fix(config): resolve effective subagent model
* fix(subagents): defer app config loading
* fix(subagents): fully defer config.yaml load in executor __init__
The previous attempt only relocated the explicit get_app_config() call,
but left resolve_subagent_model_name(...) running eagerly in __init__.
That helper has its own internal get_app_config() fallback, which still
fired when both app_config and parent_model were None and
config.model == "inherit" — exactly the path unit tests hit, breaking
21 tests in CI with FileNotFoundError: config.yaml.
Skip the eager resolve in __init__ when it would require loading the
config file, and defer to _create_agent (which already has the
app_config or get_app_config() fallback).
* fix(harness): resolve runtime paths from project root
* docs(config): update
* fix(config): address runtime path review feedback
* test(config): fix skills path e2e root
* test(config): cover legacy config fallback when project root lacks config files
Verifies that when DEER_FLOW_PROJECT_ROOT is unset and cwd has no
config.yaml/extensions_config.json, AppConfig and ExtensionsConfig fall back
to the legacy backend/repo-root candidates — the backward-compat path
requested in PR #2642 review.
---------
Co-authored-by: Willem Jiang <willem.jiang@gmail.com>
* fix(agents): propagate agent_name into ToolRuntime.context for setup_agent (#2677)
When creating a custom agent via the web UI, SOUL.md was always written
to the global base_dir/SOUL.md instead of agents/<name>/SOUL.md.
Root cause: the bootstrap flow sends agent_name via body.context, but
two layers were broken:
1. services.py only forwarded body.context keys into config["configurable"];
config["context"] was never populated.
2. worker.py constructed the parent Runtime with a hard-coded
{thread_id, run_id} context, ignoring config["context"] entirely.
After the langgraph >= 1.1.9 bump (#98a5b34f), ToolRuntime.context no
longer falls back to configurable, so setup_agent's
runtime.context.get("agent_name") returned None and the tool's silent
agent_name=None -> base_dir fallback kicked in, overwriting the global
SOUL.md.
Fix:
- services.py: extract merge_run_context_overrides() and write the
whitelisted context keys into both configurable (legacy readers) and
context (langgraph 1.1+ ToolRuntime consumers).
- worker.py: extract _build_runtime_context() and merge config["context"]
into the Runtime's context (without letting callers override
thread_id/run_id).
The base_dir fallback in setup_agent_tool.py is left in place because
the IM /bootstrap channel command depends on it. That code path can
be tightened in a follow-up.
Adds regression tests covering both helpers.
* Apply suggestions from code review
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
---------
Co-authored-by: Willem Jiang <willem.jiang@gmail.com>
Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com>
Centralize log level parsing in `logging_level_from_config()` and
application in `apply_logging_level()` within `deerflow.config.app_config`.
- Gateway lifespan applies configured log level on startup
- `debug.py` uses shared helpers instead of local duplicates
- `apply_logging_level()` targets only `deerflow`/`app` logger hierarchies
so third-party library verbosity is not affected; root handler levels
are only lowered (never raised) to allow configured loggers through
without suppressing third-party output; root logger level is not modified
- Config field description updated to clarify scope
- Tests save/restore global logging state to avoid test pollution
Co-authored-by: Claude Opus 4.7 <noreply@anthropic.com>
* fix(memory): replace short-lived asyncio.run() with persistent event loop to prevent zombie httpx connections
The memory updater used asyncio.run() inside daemon threads, creating
and destroying short-lived event loops on every update. Langchain
providers (e.g. langchain-anthropic) cache httpx AsyncClient instances
globally via @lru_cache, so SSL connections created on a loop that is
subsequently destroyed become zombie connections in the shared pool.
When the main agent's lead run later reuses one of these connections,
httpx/anyio triggers RuntimeError: Event loop is closed during
connection cleanup.
Replace the ThreadPoolExecutor + asyncio.run() pattern with a
_MemoryLoopRunner that maintains a single persistent event loop in a
daemon thread for the process lifetime. Since the loop never closes,
connections bound to it never become invalid. The _run_async_update_sync
function now submits coroutines to this persistent loop via
run_coroutine_threadsafe instead of creating throwaway loops.
* update the code to address the review comments
* Fix the review comments of 2615
P1 — user_id forwarded through sync path: Added user_id parameter to _prepare_update_prompt, _finalize_update, and _do_update_memory_sync, and forwarded it to get_memory_data(agent_name, user_id=user_id) and
save(..., user_id=user_id). The update_memory() entry point now passes user_id through both the executor.submit path and the direct call path. Added TestUserIdForwarding with two regression tests (sync + async)
verifying get_memory_data and save receive the correct user_id.
P2 — aupdate_memory() delegates to sync: Replaced the model.ainvoke() call with asyncio.to_thread(self._do_update_memory_sync, ...). This eliminates the unsafe async provider client path entirely — all memory
updater entry points now use the isolated sync model.invoke() path. Updated the test from asserting ainvoke is awaited to asserting invoke is called and ainvoke is not.
Nit — duplicate comment removed: Removed the duplicated # Matches sentences... comment on line 230.
* Chore(test): update the code of test_memory_updater
---------
Co-authored-by: rayhpeng <rayhpeng@gmail.com>
* refactor: thread app_config through middleware factories
Continues the incremental config-refactor sequence (#2611 root, #2612 lead
path) one layer deeper into the middleware factories. Two ambient lookups
inside _build_runtime_middlewares are eliminated and the LLMErrorHandling
band-aid removed:
- _build_runtime_middlewares / build_lead_runtime_middlewares /
build_subagent_runtime_middlewares now require app_config: AppConfig.
- get_guardrails_config() inside the factory is replaced with
app_config.guardrails (semantically identical — same default-factory
GuardrailsConfig — verified by direct equality check).
- LLMErrorHandlingMiddleware.__init__ now requires app_config and reads
circuit_breaker fields directly. The class-level
circuit_failure_threshold / circuit_recovery_timeout_sec defaults are
removed along with the try/except (FileNotFoundError, RuntimeError):
pass band-aid — the let-it-crash invariant the rest of the refactor
enforces.
Caller chain (already-resolved app_config sources):
- _build_middlewares in lead_agent/agent.py: reorder so
resolved_app_config = app_config or get_app_config() is computed BEFORE
build_lead_runtime_middlewares is called, then passed as kwarg.
- SubagentExecutor: optional app_config parameter (mirrors the lead-agent
pattern); _create_agent does the same `or get_app_config()` fallback at
agent-build time, so task_tool callers don't need to plumb app_config
through yet (typed-context plumbing for tool runtimes is a separate
refactor).
Tests:
- test_llm_error_handling_middleware: _make_app_config helper using
AppConfig(sandbox=SandboxConfig(use="test")) — same minimal-config
pattern conftest already uses. Three direct LLMErrorHandlingMiddleware()
calls each followed by post-construction circuit_breaker mutation fold
cleanly into _build_middleware(circuit_failure_threshold=...,
circuit_recovery_timeout_sec=...).
Verification:
- tests/test_llm_error_handling_middleware.py — 14 passed
- tests/test_subagent_executor.py — 28 passed
- tests/test_tool_error_handling_middleware.py — 6 passed
- tests/test_task_tool_core_logic.py — 18 passed (verifies task_tool
unchanged behavior)
- Full suite: 2697 passed, 3 skipped. The single intermittent failure in
tests/test_client_e2e.py::test_tool_call_produces_events is pre-existing
LLM flakiness (the test asserts the model decided to call a tool;
reproduces 1/3 on unchanged main as well).
* fix: address middleware app config review comments
* fix: satisfy app config annotation lint
* test: cover explicit app config middleware wiring
---------
Co-authored-by: greatmengqi <chenmengqi.0376@bytedance.com>
* feat(models): 适配 MindIE引擎的模型
* test: add unit tests for MindIEChatModel adapter and fix PR review comments
* chore: update uv.lock with pytest-asyncio
* build: add pytest-asyncio to test dependencies
* fix: address PR review comments (lazy import, cache clients, safe newline escape, strict xml regex)
* fix(mindie): preserve string args without JSON quotes in XML tool call serialization
* fix(mindie): preserve string args without JSON quotes in XML tool call serialization
* test_mindie_provider:format
* Potential fix for pull request finding
Co-authored-by: Copilot Autofix powered by AI <175728472+Copilot@users.noreply.github.com>
* fix(mindie): prevent nested tool_call params from leaking into outer args
* fixed by escaping XML entities in _fix_messages and unescaping during parse, with regression tests added.
---------
Co-authored-by: Willem Jiang <willem.jiang@gmail.com>
Co-authored-by: Copilot Autofix powered by AI <175728472+Copilot@users.noreply.github.com>
1445043649