bd-generate-prd
Pass
Audited by Gen Agent Trust Hub on Apr 22, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by ingesting untrusted data from the local repository and user interviews to generate output.
- Ingestion points: Data is gathered from the local codebase (via repository exploration) and user-provided feature descriptions or interview responses.
- Boundary markers: Absent. The instructions do not provide delimiters or warnings to the agent to ignore instructions found within the processed data.
- Capability inventory: The skill can read and write to the filesystem (under the .plans directory) and perform network operations by interacting with the GitHub API to submit issues.
- Sanitization: Absent. No validation or filtering is performed on the gathered content before it is processed or submitted externally.
Audit Metadata