write-a-prd
Pass
Audited by Gen Agent Trust Hub on Sep 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill implements a strong security control by requiring a human-in-the-loop. It mandates that the agent must present the full PRD to the user and obtain explicit approval before performing a write operation to GitHub, effectively preventing automated data exfiltration or unauthorized changes.
- [INDIRECT_PROMPT_INJECTION]: The skill defines a surface for processing untrusted data which could contain embedded instructions.
- Ingestion points: Content is ingested from user interviews and repository exploration as defined in SKILL.md.
- Boundary markers: The skill contains explicit instructions for user review and approval which serve as a boundary marker for the final output.
- Capability inventory: The agent is authorized to create GitHub issues, which constitutes a write capability to an external service.
- Sanitization: The skill relies on the user's manual review of the generated PRD to identify and remove any malicious content before submission.
Audit Metadata