review
Pass
Audited by Gen Agent Trust Hub on May 15, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No malicious code, obfuscation, or unauthorized data access patterns were detected. The skill operates within its stated purpose of legal document classification and routing.
- [DATA_EXPOSURE]: The skill is designed to ingest legal documents from local paths, cloud links, or direct input. While it accesses user-provided content, it does so as a primary function for analysis and does not demonstrate any patterns of unauthorized data leakage or exfiltration.
- [PROMPT_INJECTION]: The skill processes external, untrusted content (contracts and agreements). This presents a surface for indirect prompt injection where malicious instructions could be hidden inside a document to influence the AI's review behavior. This is a common risk for all document-analysis tools and does not indicate a flaw in the skill's own code.
Audit Metadata