chore(restack): advance CodeQL handler through Noema follow-up

.github/workflows/agent-review-runtime-quality-ci.yml

@@ -11,6 +11,11 @@ on:
       - "tests/test_noema_two_phase_handoff.py"
       - "tests/test_noema_refreshed_app_identity.py"
       - "tests/test_noema_token_lifetime_stale_run_contract.py"
+      - "scripts/ci/noema_review_document.py"
+      - "scripts/ci/noema_hwp_mcp_reader.mjs"
+      - "scripts/ci/noema-document-reader/package.json"
+      - "scripts/ci/noema-document-reader/package-lock.json"
+      - "tests/test_noema_document_review_context.py"
       - "docs/doctoring/noema-review-token-lifetime.md"
       - "docs/product-technical-gap-baseline.md"
       - ".github/workflows/opencode-review-dispatch.yml"
@@ -185,6 +190,11 @@ jobs:
               tests/test_noema_two_phase_handoff.py|\
               tests/test_noema_refreshed_app_identity.py|\
               tests/test_noema_token_lifetime_stale_run_contract.py|\
+              scripts/ci/noema_review_document.py|\
+              scripts/ci/noema_hwp_mcp_reader.mjs|\
+              scripts/ci/noema-document-reader/package.json|\
+              scripts/ci/noema-document-reader/package-lock.json|\
+              tests/test_noema_document_review_context.py|\
               docs/doctoring/noema-review-token-lifetime.md)
                 noema_suite=true
                 ;;
@@ -349,13 +359,15 @@ jobs:
             tests/test_noema_reviewer_token_lifetime.py \
             tests/test_noema_two_phase_handoff.py \
             tests/test_noema_refreshed_app_identity.py \
-            tests/test_noema_token_lifetime_stale_run_contract.py
+            tests/test_noema_token_lifetime_stale_run_contract.py \
+            tests/test_noema_document_review_context.py
           python -m compileall -q \
             .github/actions/noema-review/two_phase.py \
             tests/test_noema_reviewer_token_lifetime.py \
             tests/test_noema_two_phase_handoff.py \
             tests/test_noema_refreshed_app_identity.py \
-            tests/test_noema_token_lifetime_stale_run_contract.py
+            tests/test_noema_token_lifetime_stale_run_contract.py \
+            tests/test_noema_document_review_context.py
 
       - name: Verify OpenCode Rust coverage toolchain contract
         if: steps.affected_suites.outputs.opencode == 'true'

scripts/ci/noema_hwp_mcp_reader.mjs

@@ -28,7 +28,10 @@ try {
   const document = await documentModule.openDocument(filePath);
   documentModule.closeDocument(document);
   const text = await toolsModule.readHwp({ file_path: filePath });
-  if (!text || /(?:파일 읽기 오류|File not found|text extraction error)/i.test(text)) {
+  if (
+    !text ||
+    /^(?:파일 읽기 오류|File not found|텍스트 추출 오류|text extraction error)/i.test(text)
+  ) {
     throw new Error("hwp-mcp returned an extraction error");
   }
   process.stdout.write(`${text}\n`);

scripts/ci/noema_review_document.py

@@ -16,6 +16,7 @@
 from pathlib import PurePosixPath
 
 from defusedxml import ElementTree as ET
+from defusedxml.common import DefusedXmlException
 
 
 MAX_DOCUMENT_BYTES = 8 * 1024 * 1024
@@ -79,7 +80,7 @@ def _extract_docx(raw: bytes) -> str:
 
     try:
         root = ET.fromstring(document_xml)
-    except ET.ParseError as exc:
+    except (ET.ParseError, DefusedXmlException) as exc:
         raise DocumentReadError("DOCX document.xml is malformed") from exc
 
     body = root.find(f"{W}body")

tests/test_noema_document_review_context.py

@@ -34,6 +34,19 @@ def _docx_bytes(*, malformed: bool = False) -> bytes:
     return output.getvalue()
 
 
+def _docx_entity_bytes() -> bytes:
+    """Build a DOCX whose entity declaration must be rejected safely."""
+    xml = """<?xml version="1.0" encoding="UTF-8"?>
+<!DOCTYPE w:document [<!ENTITY expansion "blocked">]>
+<w:document xmlns:w="http://schemas.openxmlformats.org/wordprocessingml/2006/main">
+  <w:body><w:p><w:r><w:t>&expansion;</w:t></w:r></w:p></w:body>
+</w:document>"""
+    output = io.BytesIO()
+    with zipfile.ZipFile(output, "w", zipfile.ZIP_DEFLATED) as archive:
+        archive.writestr("word/document.xml", xml)
+    return output.getvalue()
+
+
 def _pr() -> dict[str, object]:
     return {
         "headRefOid": "head",
@@ -75,6 +88,15 @@ def test_hosted_reader_bundle_is_pinned_and_local():
     assert "python3 -m pip install --quiet --require-hashes --no-deps" in workflow
     assert "requirements-noema-document-ci-hashes.txt" in quality_workflow
     assert "Install exact Noema document dependencies" in quality_workflow
+    for path in (
+        "scripts/ci/noema_review_document.py",
+        "scripts/ci/noema_hwp_mcp_reader.mjs",
+        "scripts/ci/noema-document-reader/package.json",
+        "scripts/ci/noema-document-reader/package-lock.json",
+        "tests/test_noema_document_review_context.py",
+    ):
+        assert path in quality_workflow
+    assert "tests/test_noema_document_review_context.py" in quality_workflow
 
 
 def test_docx_text_reaches_the_actual_reviewer_payload(monkeypatch):
@@ -148,6 +170,12 @@ def test_malformed_docx_is_explicit_in_review_context(monkeypatch):
     assert "not a zip archive" not in context
 
 
+def test_forbidden_docx_entities_are_explicitly_rejected():
+    """Defused XML entity failures become the same bounded reader error."""
+    with pytest.raises(document.DocumentReadError, match="DOCX document.xml is malformed"):
+        document.extract_review_document("docs/entity.docx", _docx_entity_bytes())
+
+
 def test_hwp_reader_contract_is_local_and_fail_closed(monkeypatch):
     """HWP/HWPX use the configured local adapter and reject failed readers."""
     monkeypatch.setenv(document.HWP_READER_ENV, "/trusted/hwp-mcp-source")