Files
roboco/tests/unit/gateway/test_audit_on_rejection.py
T
Renn F 4829f93a68 fix(gateway): unblock task claim; full Phase 0/1/2 remediation
Resolves the 100% claim-failure rate introduced by the gateway rewrite
  (commit 62bda0c plus 78 follow-ups). Live smoke runs hit
  `404 /api/v2/flow/developer/...` on every dev verb plus a manifest
  fallback that silently exposed off-role verbs to PMs — confirmed
  firing simultaneously in NAS agent logs (be-dev-1, be-pm, main-pm).

  Audit reports under docs/internal/audit_2026_05_04/ catalogue 49
  defects across gateway, services, prompts, MCP transport, substrate,
  and tests (8 detail reports + master synthesis). Six smoking guns;
  three proven in production logs.

  Phase 0 — unblock claim:
  - URL prefix /api/v2/flow/dev → /developer; slug-map board roles
    (product_owner, head_marketing) → /board (D-01)
  - _i_will_work_on AttributeError on None across pending /
    needs_revision / claimed re-entry branches (D-02)
  - Seed last_heartbeat_at in _qa_or_doc_claim (D-03)
  - Drop misleading i_have_committed verb; dev flow uses commit() (D-04)
  - Manifest mount via compose; flow_server + do_server fail loud
    instead of exposing all-verbs fallback (D-12)
  - MCP _post() surfaces envelope body on 4xx so agents see remediate
    hints (D-13)
    on git failure so retries aren't blocked by half-state (S-01)

  Phase 1 — lifecycle stability:
  - _resolve_skill falls back to AgentTable.capabilities (D-06)
  - main_pm_complete uses kwargs for escalate_to_ceo (D-07)
  - i_am_done auto-runs submit_verification when in_progress (D-08)
  - active_claimant_id wired in claim/unclaim paths — single-claimant
    invariant now functional (D-05)
  - qa_pass/qa_fail assert claimed_by parity with qa_agent_id (D-18)
  - Prompt-drift sweep: fail() shape, i_am_done(task_id, notes),
    subtask cap (12 hard / 8 soft), error-code symbology rewritten in
    base.md + per-role anti-patterns (D-10/11/29/30/31, D-37)

  Phase 2 — invariants + architecture:
  - Real-DB integration test exercising claim → in_progress → commit
    → submit_for_qa → i_am_done → awaiting_qa (P2-1)
  - choreographer.py → package; 3 of 6 role mixins extracted
    (board, doc, qa). _impl.py 2,526 → 2,080 lines (-18%). Continuation
    plan in docs/internal/audit_2026_05_04/p2_2_decompose_plan.md (P2-2)
  - Closure guards consolidated via _subtasks_not_terminal_envelope (P2-3)
  - TaskService.unclaim_for_reaper routed through canonical
    _validate_and_set_status; in_progress → pending added to
    VALID_TRANSITIONS (P2-4)
  - Dead code removed: i_am_done_with_catchup verb, _run_catch_up helper
    (P2-5)
  - 6 state-machine invariants asserted via property test (P2-6)
  - attempt_id (uuid4) stamped on every gateway.rejected audit row (P2-7)
  - _reconcile_orphan_claims_on_startup rolls back tasks left CLAIMED
    with branch_name=NULL from prior crashes (P2-8)
  - scripts/regenerate_verb_tables.py introspects Pydantic schemas +
    role_config; compose_prompt injects per-role tables as a layer.
    Eliminates the prompt-drift class structurally (P2-9)

  Other:
  - D-48: orchestrator mounts host's ~/.claude.json when present so
    agents don't boot from backup recovery on every spawn
  - D-49: dev dispatcher rejects role-mismatched spawns (e.g. doc task
    assigned to dev agent)

  Tests: 553 pass · ruff + mypy clean. Live NAS smoke verification
  pending — needs the stack brought back up.
2026-05-04 23:43:55 +02:00

171 lines
5.8 KiB
Python

"""Every Envelope rejection from a Choreographer verb writes an audit row.
Choreographer takes an ``audit`` dependency but historically never invoked
it. The result: every rejection envelope (invalid_state, not_authorized,
tracing_gap, not_found) silently disappeared. With no forensic trail, a
stuck flow had no breadcrumbs.
These tests pin the ``gateway.rejected`` audit-write behavior across the
range of rejection-returning verbs, including:
- not_authorized rejections (PM cannot execute code, role-typed claim)
- invalid_state rejections (no active task, expected status mismatch)
- tracing_gap rejections (missing notes, missing journal entries)
- not_found rejections (unknown task id)
The audit call is fire-and-forget; an exception inside ``log_event`` must
not propagate or alter the envelope returned to the agent.
"""
from __future__ import annotations
from typing import Any
from unittest.mock import AsyncMock, MagicMock
from uuid import uuid4
import pytest
from roboco.services.gateway.choreographer import Choreographer, ChoreographerDeps
def _make_deps(**overrides: Any) -> ChoreographerDeps:
base: dict[str, Any] = {
"task": AsyncMock(),
"work_session": AsyncMock(),
"git": AsyncMock(),
"a2a": AsyncMock(),
"journal": AsyncMock(),
"audit": AsyncMock(),
"evidence_repo": AsyncMock(),
}
base.update(overrides)
repo = base["evidence_repo"]
for method in (
"list_unread_a2a",
"list_unread_mentions",
"list_pending_notifications",
"task_metadata_gaps",
"recent_team_activity",
"blockers_in_lane",
"journal_highlights_for_task",
):
getattr(repo, method).return_value = []
return ChoreographerDeps(**base)
# ---------------------------------------------------------------------------
# Primary acceptance test: PM cannot claim a code task — not_authorized path
# ---------------------------------------------------------------------------
@pytest.mark.asyncio
async def test_pm_cannot_execute_code_writes_audit_row() -> None:
"""A cell_pm calling i_will_work_on on a code task must:
1. Return an Envelope with error == 'not_authorized'
2. Write a gateway.rejected audit event with verb + reason details
"""
aid = uuid4()
tid = uuid4()
code_task = MagicMock(
id=tid,
status="pending",
assigned_to=aid,
task_type="code",
priority=1,
parent_task_id=None,
sequence=0,
team="backend",
)
task_svc = AsyncMock()
task_svc.get.return_value = code_task
task_svc.agent_for.return_value = MagicMock(role="cell_pm", team="backend")
task_svc.list_in_progress_for_agent.return_value = []
task_svc.list_paused_for_agent.return_value = []
task_svc.get_subtasks.return_value = []
audit_svc = AsyncMock()
deps = _make_deps(task=task_svc, audit=audit_svc)
c = Choreographer(deps)
env = await c.i_will_work_on(aid, tid, plan="x")
assert env.error == "not_authorized"
audit_svc.log_event.assert_awaited()
args = audit_svc.log_event.await_args
assert args.kwargs["event_type"] == "gateway.rejected"
assert args.kwargs["details"]["verb"] == "i_will_work_on"
assert args.kwargs["details"]["reason"] == "not_authorized"
# ---------------------------------------------------------------------------
# not_found path: unknown task id
# ---------------------------------------------------------------------------
@pytest.mark.asyncio
async def test_unknown_task_writes_audit_row() -> None:
"""not_found rejection (unknown task id) is audited."""
aid = uuid4()
tid = uuid4()
task_svc = AsyncMock()
task_svc.get.return_value = None
audit_svc = AsyncMock()
deps = _make_deps(task=task_svc, audit=audit_svc)
c = Choreographer(deps)
env = await c.i_am_done(aid, tid, notes="something")
assert env.error == "not_found"
audit_svc.log_event.assert_awaited()
args = audit_svc.log_event.await_args
assert args.kwargs["event_type"] == "gateway.rejected"
assert args.kwargs["details"]["verb"] == "i_am_done"
assert args.kwargs["details"]["reason"] == "not_found"
# ---------------------------------------------------------------------------
# Happy path must NOT write an audit row.
# ---------------------------------------------------------------------------
@pytest.mark.asyncio
async def test_successful_verb_does_not_write_audit_row() -> None:
"""Successful (non-error) Envelope must not emit gateway.rejected audit."""
aid = uuid4()
task_svc = AsyncMock()
task_svc.list_assigned_for_agent.return_value = []
task_svc.list_paused_for_agent.return_value = []
audit_svc = AsyncMock()
deps = _make_deps(task=task_svc, audit=audit_svc)
c = Choreographer(deps)
env = await c.give_me_work(aid)
# No rejection, so no audit row.
assert env.error is None
audit_svc.log_event.assert_not_awaited()
# ---------------------------------------------------------------------------
# Audit failure must NOT block the verb (best-effort rule).
# ---------------------------------------------------------------------------
@pytest.mark.asyncio
async def test_audit_log_event_failure_does_not_propagate() -> None:
"""If log_event raises, the verb still returns the rejection envelope."""
aid = uuid4()
tid = uuid4()
task_svc = AsyncMock()
# Unknown task id triggers not_found rejection on i_am_done.
task_svc.get.return_value = None
audit_svc = AsyncMock()
audit_svc.log_event.side_effect = RuntimeError("audit DB down")
deps = _make_deps(task=task_svc, audit=audit_svc)
c = Choreographer(deps)
# Must not raise; the rejection envelope should still come back.
env = await c.i_am_done(aid, tid, notes="x")
assert env.error == "not_found"
audit_svc.log_event.assert_awaited()