write-a-prd

Pass

Audited by Gen Agent Trust Hub on Aug 26, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill facilitates documentation creation by interacting with the user and exploring the repository. It uses GitHub, a well-known service, to post the resulting PRD as an issue, which aligns with its stated purpose.
  • [PROMPT_INJECTION]: The skill has an ingestion surface for untrusted data from the codebase and user input, which is used to generate the PRD. However, the skill lacks high-risk capabilities like arbitrary command execution or private file exfiltration.
  • Ingestion points: Codebase exploration (SKILL.md, Step 2) and user interview (SKILL.md, Steps 1 and 3).
  • Boundary markers: None explicitly defined in the instructions.
  • Capability inventory: Repository read access and GitHub issue creation via network write.
  • Sanitization: The skill enforces a specific structural template for the final output.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 26, 2026, 04:25 PM
Security Audit — agent-trust-hub — write-a-prd