Merge remote-tracking branch 'origin/main' into codex/aiauditbridge-single-reviewer-cleanup-20260729

Co-Authored-By: Codex <noreply@openai.com>

# Conflicts:
#	.github/codex_auto_merge_policy.json

Author: Pigbibi

scripts/gate_codex_app_review_static.py

@@ -7,7 +7,7 @@
 
 import json
 import re
-from pathlib import Path
+from pathlib import Path, PurePosixPath
 from typing import Any
 
 DEFAULT_POLICY_PATH = Path(".github/codex_auto_merge_policy.json")
@@ -70,16 +70,72 @@ def scan_diff(diff_text: str, path_patterns: list[re.Pattern[str]]) -> list[str]
     return list(dict.fromkeys(violations))
 
 
+def _safe_exact_paths(values: Any) -> set[str] | None:
+    if not isinstance(values, list) or not values:
+        return None
+    paths: set[str] = set()
+    for value in values:
+        if not isinstance(value, str) or value != value.strip():
+            return None
+        path = PurePosixPath(value)
+        if (
+            not value
+            or path.is_absolute()
+            or path.as_posix() != value
+            or ".." in path.parts
+            or "\\" in value
+            or any(character in value for character in "*?[]")
+            or value in paths
+        ):
+            return None
+        paths.add(value)
+    return paths
+
+
 def check_metadata(files: list[dict[str, Any]], policy: dict[str, Any]) -> list[str]:
     issues: list[str] = []
+    approved_deleted_paths: set[str] = set()
     max_files = policy.get("max_changed_files", 50)
     max_lines = policy.get("max_changed_lines", 5000)
+    changed_paths = {
+        filename
+        for f in files
+        if isinstance((filename := f.get("filename")), str)
+    }
+    removed_paths = {
+        f["filename"]
+        for f in files
+        if isinstance(f.get("filename"), str)
+        and (f.get("status") or "").lower().strip() == "removed"
+    }
+    configured_bundles = policy.get("approved_change_bundles", [])
+    if isinstance(configured_bundles, list):
+        for bundle in configured_bundles:
+            if not isinstance(bundle, dict):
+                continue
+            exact_changed_paths = _safe_exact_paths(bundle.get("exact_changed_paths"))
+            exact_deleted_paths = _safe_exact_paths(bundle.get("exact_deleted_paths"))
+            bundle_max_lines = bundle.get("max_changed_lines")
+            if (
+                exact_changed_paths is None
+                or exact_deleted_paths is None
+                or not exact_deleted_paths.issubset(exact_changed_paths)
+                or type(bundle_max_lines) is not int
+                or bundle_max_lines < max_lines
+                or changed_paths != exact_changed_paths
+                or removed_paths != exact_deleted_paths
+            ):
+                continue
+            approved_deleted_paths = exact_deleted_paths
+            max_lines = bundle_max_lines
+            break
+
     total_added = sum(f.get("additions", 0) or 0 for f in files)
     total_deleted = sum(f.get("deletions", 0) or 0 for f in files)
     for f in files:
         filename = f.get("filename", "?")
         status = (f.get("status") or "").lower().strip()
-        if status == "removed":
+        if status == "removed" and filename not in approved_deleted_paths:
             issues.append(f"**File deleted**: `{filename}` — verify intentional")
         elif status == "renamed":
             issues.append(f"**File renamed**: `{f.get('previous_filename', '?')}` → `{filename}`")

tests/test_gate_codex_app_review.py

@@ -107,6 +107,99 @@ def test_collect_static_gate_issues_aggregates_metadata_and_diff(self) -> None:
         self.assertTrue(any("Hardcoded secret" in issue for issue in issues))
         self.assertTrue(any("Too many lines" in issue for issue in issues))
 
+    def test_check_metadata_allows_only_an_exact_base_approved_change_bundle(self) -> None:
+        files = [
+            {
+                "filename": "scripts/retired.py",
+                "status": "removed",
+                "additions": 0,
+                "deletions": 150,
+            },
+            {
+                "filename": "service/caller.py",
+                "status": "modified",
+                "additions": 2_400,
+                "deletions": 0,
+            },
+        ]
+        policy = {
+            "max_changed_files": 10,
+            "max_changed_lines": 2_000,
+            "approved_change_bundles": [
+                {
+                    "exact_changed_paths": [
+                        "scripts/retired.py",
+                        "service/caller.py",
+                    ],
+                    "exact_deleted_paths": ["scripts/retired.py"],
+                    "max_changed_lines": 3_000,
+                }
+            ],
+        }
+
+        issues = gate_codex_app_review_static.check_metadata(files, policy)
+
+        self.assertEqual(issues, [])
+
+    def test_check_metadata_does_not_expand_line_budget_for_an_unrelated_change(self) -> None:
+        files = [
+            {
+                "filename": "service/unrelated.py",
+                "status": "modified",
+                "additions": 2_500,
+                "deletions": 0,
+            },
+        ]
+        policy = {
+            "max_changed_files": 10,
+            "max_changed_lines": 2_000,
+            "approved_change_bundles": [
+                {
+                    "exact_changed_paths": [
+                        "scripts/retired.py",
+                        "service/caller.py",
+                    ],
+                    "exact_deleted_paths": ["scripts/retired.py"],
+                    "max_changed_lines": 3_000,
+                }
+            ],
+        }
+
+        issues = gate_codex_app_review_static.check_metadata(files, policy)
+
+        self.assertTrue(any("Too many lines" in issue for issue in issues))
+
+    def test_check_metadata_rejects_partial_or_unsafe_change_bundles(self) -> None:
+        files = [
+            {
+                "filename": "scripts/retired.py",
+                "status": "removed",
+                "additions": 0,
+                "deletions": 10,
+            },
+        ]
+        for exact_changed_paths in (
+            ["scripts/retired.py", "service/caller.py"],
+            ["scripts/*.py"],
+            ["../scripts/retired.py"],
+            ["/scripts/retired.py"],
+            ["scripts\\retired.py"],
+        ):
+            with self.subTest(exact_changed_paths=exact_changed_paths):
+                policy = {
+                    "max_changed_files": 10,
+                    "max_changed_lines": 100,
+                    "approved_change_bundles": [
+                        {
+                            "exact_changed_paths": exact_changed_paths,
+                            "exact_deleted_paths": ["scripts/retired.py"],
+                            "max_changed_lines": 200,
+                        }
+                    ],
+                }
+                issues = gate_codex_app_review_static.check_metadata(files, policy)
+                self.assertTrue(any("scripts/retired.py" in issue for issue in issues))
+
 
 if __name__ == "__main__":
     unittest.main()

tests/test_single_pr_reviewer_contract.py

@@ -43,6 +43,8 @@ def test_retired_pr_reviewer_is_not_advertised_as_active() -> None:
         (ROOT / ".github/codex_auto_merge_policy.json").read_text(encoding="utf-8")
     )
     assert "pr_review" not in policy
+    assert "approved_change_bundles" not in policy
+    assert policy["max_changed_lines"] == 2_000
     assert "Codex PR Review" not in DEFAULT_WORKFLOW_ALLOWLIST
 
     for relative_path in (