to-prd-pro
Pass
Audited by Gen Agent Trust Hub on Oct 16, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill leverages the GitHub CLI (
gh) to create issues on behalf of the user. This is an expected and documented functionality using a well-known service tool. - [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from the local codebase and conversation history. While this is an inherent surface for indirect prompt injection, the risk is managed by the use of a predefined PRD template and instructions that discourage the reproduction of raw codebase content.
- [SAFE]: No signs of obfuscation, hardcoded credentials, unauthorized persistence, or malicious exfiltration were detected. The skill's behavior is consistent with its stated purpose of professional product documentation.
Audit Metadata