doc-intel-contract-pl

Pass

Audited by Gen Agent Trust Hub on Aug 17, 2026

Risk Level: SAFECREDENTIALS_UNSAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [CREDENTIALS_UNSAFE]: Hardcoded dummy API key patterns (e.g., 'sk-live-...', 'ghp_...') are present in tests/test_mask_for_model.py. These are used exclusively as test cases to verify the mask_for_model.py script's ability to redact sensitive information and are not functional credentials.
  • [COMMAND_EXECUTION]: The file tests/test_encoding_crlf.py utilizes subprocess.run to call the skill's own normalize.py script. This is performed for integration testing purposes to verify character encoding handling and does not represent a vulnerability for user input injection.
  • [EXTERNAL_DOWNLOADS]: The skill utilizes the pdf-inspector package for document classification. This is an official dependency from a known provider and is used for local processing in accordance with the skill's privacy goals.
  • [SAFE]: The skill includes a 'constitution' (.matematic/konstytucja.md) that explicitly forbids cloud communication and enforces deterministic processing, aligning with RODO (GDPR) and AI Act transparency requirements.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 17, 2026, 07:32 PM
Security Audit — agent-trust-hub — doc-intel-contract-pl