context-to-prd
Pass
Audited by Gen Agent Trust Hub on Oct 13, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to execute a grep command locally to identify character encoding issues (UTF-8 replacement characters) in the generated PRD file. This is a benign diagnostic check used for quality assurance of the output.
- [INDIRECT_PROMPT_INJECTION]: The skill ingests data from the local codebase and project documentation to populate the PRD. This creates a surface where malicious instructions embedded in the code could influence the agent's output.
- Ingestion points: Reads from the local codebase and existing documents in
docs/product/. - Boundary markers: None explicitly defined to separate untrusted codebase content from instructions.
- Capability inventory: Primarily file system read and write operations within the project's documentation directory.
- Sanitization: No specific sanitization or filtering is applied to the codebase content before it is processed.
Audit Metadata