test(workspace): harden release gate automation

skills/test/lbs-cli.test.ts

@@ -3708,7 +3708,7 @@ test("MCP stdio tool-call case setups pipeline and registered MCP server", () =>
     assert.deepEqual(
       run.setup_automation.map((item: { entry: string }) => item.entry),
       [
-        "node:scripts/e2e/ensure-local-agent-pipeline.mjs --write-env",
+        "node:scripts/e2e/ensure-fake-provider-pipeline.mjs --write-env",
         "case:mcp-stdio-register",
       ],
     );
@@ -3739,8 +3739,8 @@ test("MCP stdio tool-call case setups pipeline and registered MCP server", () =>
     assert.equal(planResult.code, 0);
     const plan = JSON.parse(planResult.output);
     assert.deepEqual(plan.setup_provides_env, [
-      "LANGBOT_LOCAL_AGENT_PIPELINE_URL",
-      "LANGBOT_LOCAL_AGENT_PIPELINE_NAME",
+      "LANGBOT_FAKE_PROVIDER_PIPELINE_URL",
+      "LANGBOT_FAKE_PROVIDER_PIPELINE_NAME",
       "LANGBOT_MCP_QA_STDIO_SERVER_UUID",
     ]);
     assert.ok(
@@ -4588,6 +4588,47 @@ test("test report classifies model route failures as env_issue", () => {
   }
 });
 
+test("test report classifies provider quota tracebacks as env_issue", () => {
+  const tmp = mkdtempSync(join(tmpdir(), "lbs-report-quota-env-issue-"));
+  try {
+    const logPath = join(tmp, "backend.log");
+    writeFileSync(
+      logPath,
+      [
+        "[05-21 10:31:00.000] chat.py (2) - [ERROR] : Request Failed: Traceback (most recent call last):",
+        "  File \"provider.py\", line 1, in invoke",
+        "openai.PermissionDeniedError: insufficient user quota",
+        "[05-21 10:31:01.000] pipeline.py (3) - [ERROR] : runner.llm_error All models failed during streaming setup: insufficient user quota",
+      ].join("\n"),
+    );
+
+    const result = capture(() =>
+      commandTestReport(
+        ctx([
+          "test",
+          "report",
+          "local-agent-complex-coding-task-debug-chat",
+          "--backend-log",
+          logPath,
+          "--json",
+        ]),
+      ),
+    );
+    assert.equal(result.code, 0);
+    const report = JSON.parse(result.output);
+    assert.equal(report.log_guard.status, "env_issue");
+    assert.ok(
+      report.log_guard.findings.some(
+        (finding: { severity?: string; pattern?: string }) =>
+          finding.severity === "env_issue" &&
+          finding.pattern === "insufficient user quota",
+      ),
+    );
+  } finally {
+    rmSync(tmp, { recursive: true, force: true });
+  }
+});
+
 test("test report infers scan window from automation result evidence", () => {
   const tmp = mkdtempSync(join(tmpdir(), "lbs-report-evidence-window-"));
   try {

skills/test/test_agent_run_ledger_policy.py

@@ -0,0 +1,69 @@
+from __future__ import annotations
+
+import sys
+import unittest
+from pathlib import Path
+
+
+sys.path.insert(0, str(Path(__file__).resolve().parents[1] / "scripts" / "e2e"))
+
+from agent_run_ledger_policy import (  # noqa: E402
+    classify_invalid_tool_argument_errors,
+    load_ledger_json,
+)
+
+
+class AgentRunLedgerPolicyTests(unittest.TestCase):
+    def setUp(self) -> None:
+        self.event = {
+            "sequence": 10,
+            "type": "tool.call.completed",
+            "signal": "Invalid JSON arguments",
+        }
+
+    def test_completed_run_with_later_success_is_warning(self) -> None:
+        failures, warnings = classify_invalid_tool_argument_errors(
+            [self.event],
+            successful_tool_completion_sequences=[12],
+            run_completed=True,
+        )
+
+        self.assertEqual(failures, [])
+        self.assertEqual(warnings[0]["kind"], "recovered_tool_argument_error")
+
+    def test_completed_run_without_later_success_still_fails(self) -> None:
+        failures, warnings = classify_invalid_tool_argument_errors(
+            [self.event],
+            successful_tool_completion_sequences=[8],
+            run_completed=True,
+        )
+
+        self.assertEqual(failures, [self.event])
+        self.assertEqual(warnings, [])
+
+    def test_incomplete_run_still_fails_even_with_later_success(self) -> None:
+        failures, warnings = classify_invalid_tool_argument_errors(
+            [self.event],
+            successful_tool_completion_sequences=[12],
+            run_completed=False,
+        )
+
+        self.assertEqual(failures, [self.event])
+        self.assertEqual(warnings, [])
+
+    def test_malformed_persisted_event_json_remains_an_invariant_failure(self) -> None:
+        failures: list[dict] = []
+
+        value = load_ledger_json(
+            '{"tool_call_id":',
+            field="agent_run_event[10].data_json",
+            failures=failures,
+        )
+
+        self.assertEqual(value, {})
+        self.assertEqual(failures[0]["kind"], "invalid_json")
+        self.assertEqual(failures[0]["field"], "agent_run_event[10].data_json")
+
+
+if __name__ == "__main__":
+    unittest.main()