prd-refine

Pass

Audited by Gen Agent Trust Hub on May 10, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill's functionality is consistent with its description. It reads local guide files and processes project documentation without unauthorized file access or network communication.
  • [COMMAND_EXECUTION]: The skill executes a Python-based command dispatcher (prd-command-dispatch.py) located within the skill's own directory structure. The command parameters are limited to the project root and a specific documentation directory.
  • [PROMPT_INJECTION]: The instructions do not attempt to bypass agent safety filters or override system constraints. It includes standard instructions for multi-language support.
  • [DATA_EXPOSURE]: The skill accesses documentation files in modules/refine/ and writes to docs/prd-helper/. These operations are well-defined and scoped to the project's PRD refinement task.
Audit Metadata
Risk Level
SAFE
Analyzed
May 10, 2026, 04:23 AM
Security Audit — agent-trust-hub — prd-refine