fix(company): terminalize report-chain failure into FAILED settlement; prompt-spill hygiene

Report chain: when consecutive report cards exceed the failure limit the parent now fails through transition_work_item (settlement, dependents, manager visibility) instead of parking forever in AWAITING_MANAGER_REVIEW behind a metadata hold no reconcile path could clear. The hold stamp survives only as the quarantine fallback when the FAILED write does not land, mirroring the attempt ledger's terminalize pattern. Cursor prompt spill: one stable file per task (retries overwrite instead of accumulating) and a self-ignoring .gitignore so workspace git never picks up .opc/external_prompts. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
2026-07-31 19:53:13 +08:00
parent 18ddc18948
commit 5f5fbf432e
4 changed files with 121 additions and 17 deletions
@@ -3612,6 +3612,17 @@ class ExternalAgentMonitoringTests(unittest.IsolatedAsyncioTestCase):
            self.assertIn(str(prompt_file), str(cmd[-1]))
            self.assertNotIn(prompt[:64], metadata["command"])
            self.assertLess(len(cmd[-1].encode("utf-8")), CursorAdapter._ARGV_PROMPT_MAX_BYTES)
+            # Hygiene: one stable file per task (retries overwrite) and a
+            # self-ignoring directory so workspace git never picks it up.
+            self.assertEqual(prompt_file.name, "cursor_task-large.md")
+            _cmd2, metadata2 = adapter.build_interactive_invocation(
+                task, workspace_path=tmpdir
+            )
+            self.assertEqual(metadata2["prompt_file"], str(prompt_file))
+            self.assertEqual(
+                sorted(p.name for p in prompt_file.parent.iterdir()),
+                [".gitignore", "cursor_task-large.md"],
+            )
        finally:
            _cleanup_test_dir(tmpdir)

@@ -13,7 +13,7 @@ from __future__ import annotations
 import tempfile
 import unittest
 from pathlib import Path
-from unittest.mock import AsyncMock, MagicMock
+from unittest.mock import AsyncMock, MagicMock, patch

 from opc.core.config import OPCConfig, RoleConfig
 from opc.core.events import EventBus
@@ -1503,6 +1503,39 @@ class ReportFailureStormBrakeTests(unittest.IsolatedAsyncioTestCase):
            )
        )
        refreshed = await self.store.get_delegation_work_item("wi-child")
+        self.assertEqual(refreshed.phase, Phase.FAILED)
+        self.assertIn("report_chain_failure", str(refreshed.blocked_reason or ""))
+        self.assertFalse(
+            str((refreshed.metadata or {}).get("report_chain_hold", "") or "").strip()
+        )
+
+    async def test_terminalize_write_failure_quarantines_via_hold(self) -> None:
+        parent = _build_child_work_item()
+        parent.phase = Phase.AWAITING_MANAGER_REVIEW
+        parent.metadata = {
+            **dict(parent.metadata or {}),
+            "review_owner_role_id": "cto",
+            "review_owner_seat_id": "seat::team::cto::cto",
+            "max_consecutive_report_failures": 3,
+        }
+        await self.store.save_delegation_work_item(parent)
+        for attempt in (1, 2, 3):
+            await self.store.save_delegation_work_item(self._failed_report(attempt))
+
+        with patch(
+            "opc.layer2_organization.company_mode.transition_work_item",
+            side_effect=RuntimeError("db write lost"),
+        ):
+            await self.executor._reconcile_missing_review_chain(
+                await self.store.list_delegation_work_items("run-1")
+            )
+        self.assertIsNone(
+            await self.store.get_delegation_work_item(
+                report_work_item_id_for_attempt("wi-child", 4)
+            )
+        )
+        refreshed = await self.store.get_delegation_work_item("wi-child")
+        self.assertEqual(refreshed.phase, Phase.AWAITING_MANAGER_REVIEW)
        self.assertEqual(
            refreshed.metadata.get("report_chain_hold"),
            "consecutive_report_failures",