fix: address audit findings — SDK plumbing, TUI bus, dead code

Critical fixes:

- ``StrixOrchestrationHooks.on_agent_start`` now finds the
  ``CaidoCapability`` via ``ctx.context['caido_capability']`` instead
  of ``agent.capabilities`` (we use plain ``Agent``, not
  ``SandboxAgent``, so the latter never existed). The session
  manager's bundle already exposes the capability; ``run_strix_scan``
  threads it through ``make_agent_context`` and ``create_agent``
  forwards it to children.

- ``run_strix_scan`` registers the ``StrixTracingProcessor`` with the
  SDK's tracing provider via ``add_trace_processor`` so SDK trace
  spans hit ``run_dir/events.jsonl`` (was previously a parallel stream
  the SDK ignored).

- ``on_llm_end`` now writes to ``Tracer.record_llm_usage`` in
  addition to ``bus.record_usage`` so the CLI/TUI stats panel sees
  real numbers instead of zeros.

- ``run_strix_scan`` accepts an externally-built ``AgentMessageBus``
  + an explicit ``model`` arg. The TUI pre-creates the bus so its
  stop and chat-input handlers can submit ``bus.send`` /
  ``bus.cancel_descendants`` coroutines onto the scan thread's loop
  via ``asyncio.run_coroutine_threadsafe`` — replacing the
  TODO-stub no-ops.

- ``model`` config now propagates root → context → child agents in
  ``create_agent`` (was hardcoded fallback).

Dead-code removal:

- Deleted the ``load_skill`` tool entirely (host module, sandbox
  module, TUI renderer, tests). The legacy implementation reached
  into a global ``_agent_instances`` registry that no longer exists;
  the post-migration stub returned ``success=True`` without
  injecting anything — pure theater. Skills are still preloaded via
  the system prompt at scan-bring-up.

- Dropped ``tenacity`` and ``xmltodict`` from
  ``[project.dependencies]`` — neither is imported anywhere
  post-migration.

- Stripped the system prompt's "use the load_skill tool" lines.

Tests: 278/278 passing. Removed two ``load_skill`` test cases and a
``test_tool_registration_modes::test_load_skill_import_...`` assertion
that exercised the deleted module.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
This commit is contained in:
0xallam
2026-04-25 10:08:35 -07:00
parent af42499b95
commit 572ef2a2af
21 changed files with 98 additions and 318 deletions
@@ -58,45 +58,5 @@ def test_sandbox_registers_sandbox_tools_but_not_non_sandbox_tools(
assert "list_requests" in names
assert "create_agent" not in names
assert "finish_scan" not in names
assert "load_skill" not in names
assert "browser_action" not in names
assert "web_search" not in names
def test_load_skill_import_does_not_register_create_agent_in_sandbox(
monkeypatch: Any,
) -> None:
monkeypatch.setenv("STRIX_SANDBOX_MODE", "true")
monkeypatch.setenv("STRIX_DISABLE_BROWSER", "true")
monkeypatch.delenv("PERPLEXITY_API_KEY", raising=False)
monkeypatch.setattr(Config, "load", classmethod(_empty_config_load))
clear_registry()
for name in list(sys.modules):
if name == "strix.tools" or name.startswith("strix.tools."):
sys.modules.pop(name, None)
load_skill_module = importlib.import_module("strix.tools.load_skill.load_skill_actions")
registry = importlib.import_module("strix.tools.registry")
names_before = set(registry.get_tool_names())
assert "load_skill" not in names_before
assert "create_agent" not in names_before
state_type = type(
"DummyState",
(),
{
"agent_id": "agent_test",
"context": {},
"update_context": lambda self, key, value: self.context.__setitem__(key, value),
},
)
result = load_skill_module.load_skill(state_type(), "nmap")
names_after = set(registry.get_tool_names())
assert "create_agent" not in names_after
# load_skill no longer reaches into the legacy _agent_instances global —
# it returns ``success=True`` with the requested skills echoed back.
assert result["success"] is True
assert result["loaded_skills"] == ["nmap"]