crg-req-to-design

Pass

Audited by Gen Agent Trust Hub on Jul 12, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill documentation lists standard, reputable Python dependencies including pandas, openpyxl, and mcp. These are expected libraries for data processing and MCP server functionality.- [DATA_EXPOSURE]: The skill reads user-supplied Excel and CSV files and writes output tables and reports to a local directory. These file operations are core to the skill's purpose and are performed using standard file I/O methods without accessing sensitive system paths.- [INDIRECT_PROMPT_INJECTION]: As the skill ingests external data (requirement tables), it possesses an indirect prompt injection surface. However, the data is parsed for signal names and frequencies using structured logic and regular expressions, and is not used to construct executable commands or unescaped prompt content, which maintains a safe execution profile.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 12, 2026, 04:02 PM
Security Audit — agent-trust-hub — crg-req-to-design