write-prd
Pass
Audited by Gen Agent Trust Hub on Sep 26, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest untrusted data from user descriptions and existing repository files to generate documentation. While this creates a surface where embedded instructions in those sources could potentially influence the agent's output, this is the intended primary purpose of the skill and represents a standard operating procedure for a documentation assistant.
- Ingestion points: User-provided descriptions (Step 1), user responses to clarifying questions (Step 3), and existing codebase files (Step 2).
- Boundary markers: No explicit delimiters or instructions are provided to the agent to ignore embedded commands in the source data.
- Capability inventory: The skill utilizes file writing capabilities (Step 5) and repository exploration (Step 2).
- Sanitization: No sanitization or validation of the ingested external content is specified before interpolation into the PRD template.
- [SAFE]: The skill uses clear, non-deceptive instructions and follows a well-defined process. It explicitly separates functional requirements from technical implementation details, which is a best practice for project documentation and reduces the risk of the agent making unauthorized technical changes.
Audit Metadata