deer-flow/backend/tests/test_deferred_tool_crosscontext.py
AochenShen99 3b6dd0a4e3
feat(subagents): extend deferred MCP tool loading to subagents (#3432)
* feat(subagents): extend deferred MCP tool loading to subagents (#3341)

Subagents now reuse the lead agent's deferred-tool path: when
tool_search.enabled, MCP tool schemas are withheld from the model and
surfaced by name in <available-deferred-tools>, fetched on demand via the
generated tool_search helper. DeferredToolFilterMiddleware deterministically
rewrites request.tools to hide the deferred schemas (the prompt section is
discovery only, not enforcement).

Consolidates the assembly into deerflow.tools.builtins.tool_search, now the
single home for both assemble_deferred_tools (centralized fail-closed guard,
replacing the lead-only private _assemble_deferred) and the relocated
get_deferred_tools_prompt_section. Shared by every build path: lead agent,
embedded client, and subagent executor.

tool_search is appended after the subagent's name-level tool policy and is
treated as infrastructure: its catalog is built from the already
policy-filtered list, so it can never surface a tool the policy denied.

Follow-up to #3370. Fixes #3341.

* test(subagents): assert the real middleware builder emits a working deferred filter (#3341)

The existing recipe test hand-constructs DeferredToolFilterMiddleware, so it
cannot catch a regression in how build_subagent_runtime_middlewares (the call
executor._create_agent actually makes) wires the deferred setup into the
filter. Add a test that sources the filter from the real builder given a real
setup and runs it through a graph: a wrong catalog hash would silently stop
promotion, a dropped filter would stop hiding — both now caught.

Running the full real middleware stack is intentionally avoided (the other
runtime middlewares need sandbox/thread infra to execute, which would make the
test flaky); their attachment + ordering before Safety stays locked in
test_tool_error_handling_middleware.py.

* test(subagents): keep executor tests config-free in CI

* chore: trigger ci

* Potential fix for pull request finding

Co-authored-by: Copilot Autofix powered by AI <175728472+Copilot@users.noreply.github.com>

---------

Co-authored-by: Willem Jiang <willem.jiang@gmail.com>
Co-authored-by: Copilot Autofix powered by AI <175728472+Copilot@users.noreply.github.com>
2026-06-08 23:17:22 +08:00

174 lines
6.8 KiB
Python

"""Regressions for the deferred-tool redesign (#3272).
- Cross-context: building the graph in one async context and running it in a
sibling context (that did NOT inherit the build context) must still hide
deferred tools. The old ContextVar implementation failed this; the closure +
graph-state implementation must pass.
- Policy leak (Finding 1): a tool removed by policy must not be searchable.
- Fail-closed (Finding 2): a wiring regression must raise, not silently leak.
- #2884 isolation: a second (subagent-style) setup build must not affect the
lead agent's middleware/promotion.
"""
import asyncio
from pathlib import Path
import pytest
from langchain.agents import create_agent
from langchain_core.language_models.fake_chat_models import GenericFakeChatModel
from langchain_core.messages import AIMessage, HumanMessage
from langchain_core.tools import tool as as_tool
from deerflow.agents.middlewares.deferred_tool_filter_middleware import DeferredToolFilterMiddleware
from deerflow.skills.tool_policy import filter_tools_by_skill_allowed_tools
from deerflow.skills.types import Skill
from deerflow.tools.builtins.tool_search import DeferredToolSetup, assemble_deferred_tools, build_deferred_tool_setup
from deerflow.tools.mcp_metadata import tag_mcp_tool
@as_tool
def active_tool(x: str) -> str:
"active"
return x
@as_tool
def mcp_secret(x: str) -> str:
"deferred mcp tool — must be hidden from bind_tools until promoted"
return x
_BOUND: list[list[str]] = []
class _RecordingModel(GenericFakeChatModel):
def bind_tools(self, tools, **kwargs):
_BOUND.append([getattr(t, "name", None) for t in tools])
return self
def _build_graph():
filtered = [active_tool, tag_mcp_tool(mcp_secret)]
setup = build_deferred_tool_setup(filtered, enabled=True)
final = [*filtered, setup.tool_search_tool]
model = _RecordingModel(messages=iter([AIMessage(content="done")] * 4))
return create_agent(
model=model,
tools=final,
middleware=[DeferredToolFilterMiddleware(setup.deferred_names, setup.catalog_hash)],
system_prompt="t",
)
async def _abuild():
return _build_graph()
def test_deferred_hidden_when_built_and_run_in_different_contexts():
"""Build in one task, run in a sibling task that did not inherit it."""
_BOUND.clear()
async def main():
graph = await asyncio.create_task(_abuild())
async def run():
await graph.ainvoke({"messages": [HumanMessage(content="hi")]})
await asyncio.create_task(run())
asyncio.run(main())
assert _BOUND, "model was never bound"
assert not any("mcp_secret" in names for names in _BOUND), f"deferred MCP tool leaked into bind_tools: {_BOUND}"
def test_policy_excluded_mcp_tool_not_in_catalog():
"""Finding 1: a tool removed by policy is not searchable/exposed."""
filtered_after_policy = [active_tool] # mcp_secret denied by skill allowed-tools
setup = build_deferred_tool_setup(filtered_after_policy, enabled=True)
assert setup.deferred_names == frozenset()
assert setup.tool_search_tool is None
def test_fail_closed_when_mcp_survives_without_setup(monkeypatch):
"""Finding 2: simulate a wiring regression and assert it fails loudly.
``assemble_deferred_tools`` references ``build_deferred_tool_setup`` as a
module global, so patch it in ``tool_search`` (its home module).
"""
monkeypatch.setattr(
"deerflow.tools.builtins.tool_search.build_deferred_tool_setup",
lambda tools, *, enabled: DeferredToolSetup(None, frozenset(), None),
)
with pytest.raises(RuntimeError, match="fail-closed"):
assemble_deferred_tools([tag_mcp_tool(mcp_secret)], enabled=True)
def test_subagent_reentry_does_not_touch_lead_state():
"""#2884: building a second (subagent) setup must not affect the lead's
middleware. With no shared registry/ContextVar, the lead middleware depends
only on its own deferred_names + the passed state."""
lead_setup = build_deferred_tool_setup([active_tool, tag_mcp_tool(mcp_secret)], enabled=True)
mw = DeferredToolFilterMiddleware(lead_setup.deferred_names, lead_setup.catalog_hash)
# Simulate a subagent build re-entering tool assembly with its own setup.
_ = build_deferred_tool_setup([tag_mcp_tool(mcp_secret)], enabled=True)
class _Req:
def __init__(self):
self.tools = [active_tool, mcp_secret]
self.state = {"promoted": {"catalog_hash": lead_setup.catalog_hash, "names": ["mcp_secret"]}}
def override(self, tools):
self.tools = tools
return self
out = mw._filter_tools(_Req())
assert {t.name for t in out.tools} == {"active_tool", "mcp_secret"} # promotion intact
def _make_skill(allowed_tools):
"""Skill carrying an explicit allowed-tools allowlist (None = legacy allow-all)."""
return Skill(
name="s",
description="d",
license="MIT",
skill_dir=Path("/tmp/s"),
skill_file=Path("/tmp/s/SKILL.md"),
relative_path=Path("s"),
category="public",
allowed_tools=allowed_tools,
enabled=True,
)
def test_policy_denied_mcp_yields_no_tool_search_end_to_end():
"""An allowlist that denies the MCP tool gates it end-to-end: after the real
policy filter no MCP tool survives, so ``assemble_deferred_tools`` adds no
tool_search (and does not fail-closed, because no MCP tool leaked through)."""
filtered = filter_tools_by_skill_allowed_tools([active_tool, tag_mcp_tool(mcp_secret)], [_make_skill(["active_tool"])])
final_tools, setup = assemble_deferred_tools(filtered, enabled=True)
assert [t.name for t in final_tools] == ["active_tool"]
assert "tool_search" not in {t.name for t in final_tools}
assert setup.deferred_names == frozenset()
def test_tool_search_appended_after_policy_but_never_exposes_denied_tool():
"""Intentional behavior change vs. upstream (Copilot review on PR #3342).
``tool_search`` is appended AFTER skill-allowlist filtering, so an allowlist
can no longer deny ``tool_search`` by name. This is safe by construction: the
tool only appears when allowed MCP tools survive the filter, and its catalog
is derived from the already policy-filtered list — so it can never expose a
tool the allowlist denied. Locks that contract so the ordering cannot regress.
"""
allowed = ["active_tool", "mcp_secret"] # permits the MCP tool, does NOT list tool_search
filtered = filter_tools_by_skill_allowed_tools([active_tool, tag_mcp_tool(mcp_secret)], [_make_skill(allowed)])
final_tools, setup = assemble_deferred_tools(filtered, enabled=True)
names = {t.name for t in final_tools}
assert "tool_search" in names # appended despite not being in the allowlist
assert setup.deferred_names == frozenset({"mcp_secret"})
assert set(setup.deferred_names) <= set(allowed) # catalog never exceeds the allowlist