Make C10 recovery qualification deterministic

tests/test_live_codex_qualification_c10.py

@@ -0,0 +1,89 @@
+from __future__ import annotations
+
+from contextlib import nullcontext
+from pathlib import Path
+import sys
+import types
+import unittest
+from unittest import mock
+
+ROOT = Path(__file__).resolve().parents[1]
+sys.path.insert(0, str(ROOT / "tools"))
+
+import live_codex_qualification_c10 as c10
+
+C10_SOURCE = ROOT / "tools" / "live_codex_qualification_c10.py"
+V7_SOURCE = ROOT / "tools" / "live_codex_qualification_harness_v7.py"
+
+
+class LiveCodexC10Tests(unittest.TestCase):
+    def test_v7_installs_deterministic_c10_runtime(self) -> None:
+        source = V7_SOURCE.read_text(encoding="utf-8")
+        self.assertIn("import live_codex_qualification_c10 as c10", source)
+        self.assertIn("c10.install(v6, _live_runner_persisted_trust_runtime)", source)
+
+    def test_c10_fixture_is_outer_harness_owned_and_product_validated(self) -> None:
+        source = C10_SOURCE.read_text(encoding="utf-8")
+        self.assertIn("v4._start_active_run(", source)
+        self.assertIn("v4._create_checkpoint(planning=planning, run_root=run_root)", source)
+        self.assertGreaterEqual(source.count("v4._checkpoint_validation(planning)"), 4)
+        self.assertIn('"fixture_prepared_by_outer_harness=true"', source)
+        self.assertNotIn("planner_prompt", source)
+
+    def test_postcompact_is_independent_of_session_start_context(self) -> None:
+        source = C10_SOURCE.read_text(encoding="utf-8")
+        self.assertIn('configured.pop("SessionStart", None)', source)
+        self.assertIn('bool(configured.get("PreCompact"))', source)
+        self.assertIn('bool(configured.get("PostCompact"))', source)
+        self.assertIn("not compact_session_start", source)
+        self.assertIn("postcompact_isolated", source)
+
+    def test_compaction_trigger_is_qualification_only(self) -> None:
+        source = C10_SOURCE.read_text(encoding="utf-8")
+        self.assertIn('compat._set_feature(text, "token_budget", "false")', source)
+        self.assertIn('"token_budget_disabled_in_isolated_fixture": True', source)
+        self.assertIn("C10_SESSION_COMPACT_LIMIT = 1_000_000", source)
+        self.assertIn("C10_COMPACT_LIMIT = 200", source)
+
+    def test_opaque_recovery_value_is_not_persisted_in_evidence(self) -> None:
+        source = C10_SOURCE.read_text(encoding="utf-8")
+        self.assertIn("proof = secrets.token_hex(16)", source)
+        self.assertIn("def _payload_summary", source)
+        self.assertIn('"opaque_recovery_value_persisted=false"', source)
+        self.assertNotIn('"model_payload": session_payload', source)
+        self.assertNotIn('"model_payload": compact_payload', source)
+
+    def test_install_routes_only_c10(self) -> None:
+        calls: list[dict[str, object]] = []
+
+        def original(**kwargs: object) -> tuple[str, bool]:
+            calls.append(dict(kwargs))
+            return "ORIGINAL", False
+
+        dummy = types.SimpleNamespace(_ORIGINAL_CAPABILITY_RUNTIME=original)
+        trust = lambda: nullcontext(Path("."))
+        with mock.patch.object(c10, "run_c10", return_value=("REPRODUCED", True)) as run_c10:
+            c10.install(dummy, trust)
+            self.assertEqual(
+                dummy._ORIGINAL_CAPABILITY_RUNTIME(capability_id="C11", sentinel=1),
+                ("ORIGINAL", False),
+            )
+            self.assertEqual(
+                dummy._ORIGINAL_CAPABILITY_RUNTIME(capability_id="C10", sentinel=2),
+                ("REPRODUCED", True),
+            )
+        self.assertEqual(calls, [{"capability_id": "C11", "sentinel": 1}])
+        run_c10.assert_called_once_with(live_trust_runtime=trust, sentinel=2)
+
+    def test_safety_boundary_is_not_weakened(self) -> None:
+        source = C10_SOURCE.read_text(encoding="utf-8")
+        self.assertNotIn("--dangerously-bypass-approvals-and-sandbox", source)
+        self.assertNotIn("danger-full-access", source)
+        self.assertNotIn("--privileged", source)
+        self.assertNotIn("SYS_ADMIN", source)
+        self.assertIn('sandbox="read-only"', source)
+        self.assertIn("base.git_snapshot", source)
+
+
+if __name__ == "__main__":
+    unittest.main()