regulatory-guardrail-checker
Pass
Audited by Gen Agent Trust Hub on May 12, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill analyzes documentation locally using keyword matching. The Python script (check.py) only utilizes standard libraries and does not include network, remote code execution, or privilege escalation capabilities.- [DATA_EXPOSURE]: The script reads local files provided via the --spec argument, which is consistent with its role as a document scanner. It does not access restricted system directories or sensitive user credentials.- [PROMPT_INJECTION]: The skill processes untrusted external documentation, creating an indirect prompt injection surface. However, the risk is minimal as the tool performs static analysis and generates a structured report. Mandatory Evidence: (1) Ingestion point: check.py reads content from the file path specified in --spec; (2) Boundary markers: None; (3) Capability inventory: Local file reading and writing via standard Python libraries; (4) Sanitization: None performed on the input text.
Audit Metadata