Add validation integrity verifier for PR-head vs diagnostic tests (#316)

Separate official PR-head validation from post-edit diagnostic runs in
reviewer reports. Block treating diagnostic passes as official validation
and require contaminated sessions to report recovery-required status.

Co-Authored-By: Claude Opus 4.8 (1M context) <[email protected]>

tests/test_llm_workflow_split.py

@@ -121,4 +121,10 @@ def test_start_issue_template_references_work_issue_workflow():
 def test_create_issue_final_report_verifier_exported():
     from review_proofs import assess_create_issue_final_report
 
-    assert callable(assess_create_issue_final_report)
+    assert callable(assess_create_issue_final_report)
+
+
+def test_validation_integrity_verifier_exported():
+    from review_proofs import assess_validation_integrity_report
+
+    assert callable(assess_validation_integrity_report)

tests/test_reviewer_validation_integrity.py

@@ -0,0 +1,130 @@
+"""Tests for PR-head vs diagnostic validation integrity verifier (#316)."""
+import sys
+import unittest
+from pathlib import Path
+
+sys.path.insert(0, str(Path(__file__).resolve().parent.parent))
+
+from reviewer_validation_integrity import assess_validation_integrity_report  # noqa: E402
+
+
+def _official_only_report() -> str:
+    return "\n".join([
+        "Official PR-head validation on unmodified worktree.",
+        "Official validation command: pytest tests/ -q",
+        "Official validation result: failed (3 failed)",
+        "Validation integrity status: failed",
+        "Worktree dirty after official validation: clean",
+        "Review decision: request_changes",
+    ])
+
+
+def _diagnostic_report() -> str:
+    return "\n".join([
+        _official_only_report(),
+        "Diagnostic local experiment — not PR-head validation",
+        "Diagnostic edit path: tests/test_example.py",
+        "File edits by reviewer: tests/test_example.py",
+        "Diagnostic command: pytest tests/test_example.py -q",
+        "Diagnostic result: passed after local fix",
+        "Diagnostic results used only for suggested fix; blocker remains PR-head failure.",
+    ])
+
+
+class TestValidationIntegrity(unittest.TestCase):
+    def test_official_only_passes(self):
+        result = assess_validation_integrity_report(
+            _official_only_report(),
+            validation_session={
+                "official_validation_ran": True,
+                "official_result": "fail",
+                "worktree_dirty_after_official": False,
+            },
+        )
+        self.assertTrue(result["proven"], result["reasons"])
+
+    def test_failing_pr_head_with_diagnostic_pass_passes(self):
+        result = assess_validation_integrity_report(
+            _diagnostic_report(),
+            validation_session={
+                "official_validation_ran": True,
+                "official_result": "fail",
+                "diagnostic_validation_ran": True,
+                "diagnostic_result": "pass",
+                "diagnostic_edits": ["tests/test_example.py"],
+                "worktree_dirty_after_official": False,
+            },
+        )
+        self.assertTrue(result["proven"], result["reasons"])
+
+    def test_post_edit_tests_as_official_blocks(self):
+        report = "\n".join([
+            "Official validation result: passed after diagnostic edit",
+            "pytest passed after local fix in validation worktree",
+        ])
+        result = assess_validation_integrity_report(
+            report,
+            validation_session={
+                "official_validation_ran": True,
+                "diagnostic_edits": ["tests/test_example.py"],
+                "diagnostic_validation_ran": True,
+            },
+        )
+        self.assertFalse(result["proven"])
+        self.assertTrue(result["block"])
+
+    def test_missing_dirty_after_validation_blocks(self):
+        report = _official_only_report().replace(
+            "Worktree dirty after official validation: clean",
+            "",
+        )
+        result = assess_validation_integrity_report(
+            report,
+            validation_session={"official_validation_ran": True, "official_result": "fail"},
+        )
+        self.assertFalse(result["proven"])
+        self.assertTrue(any("dirty" in r.lower() for r in result["reasons"]))
+
+    def test_diagnostic_without_label_blocks(self):
+        report = _diagnostic_report().replace(
+            "Diagnostic local experiment — not PR-head validation",
+            "",
+        )
+        result = assess_validation_integrity_report(
+            report,
+            validation_session={
+                "official_validation_ran": True,
+                "diagnostic_edits": ["tests/test_example.py"],
+                "diagnostic_validation_ran": True,
+            },
+        )
+        self.assertFalse(result["proven"])
+        self.assertTrue(any("diagnostic" in r.lower() for r in result["reasons"]))
+
+    def test_contaminated_requires_integrity_status(self):
+        report = "Validation worktree was edited before official validation completed."
+        result = assess_validation_integrity_report(
+            report,
+            validation_session={"contaminated": True, "official_validation_ran": True},
+        )
+        self.assertFalse(result["proven"])
+        self.assertTrue(any("contaminated" in r.lower() for r in result["reasons"]))
+
+    def test_action_log_edits_trigger_diagnostic_rules(self):
+        result = assess_validation_integrity_report(
+            _official_only_report(),
+            validation_session={"official_validation_ran": True, "official_result": "fail"},
+            action_log=[{"path": "tests/test_example.py", "kind": "file_edit", "performed": True}],
+        )
+        self.assertFalse(result["proven"])
+
+
+class TestExport(unittest.TestCase):
+    def test_review_proofs_reexport(self):
+        from review_proofs import assess_validation_integrity_report as exported
+
+        self.assertTrue(callable(exported))
+
+
+if __name__ == "__main__":
+    unittest.main()