prd-approve
Pass
Audited by Gen Agent Trust Hub on Oct 31, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes content from draft PRDs and review synthesis files, creating an attack surface where malicious instructions could be embedded in the processed data.
- Ingestion points: Reads files from
outputs/prds/*.md,outputs/prds/*-review-synthesis.md, andPRD-PIPELINE.mdas specified in Steps 1, 2, 3, and 5 of the SKILL.md. - Boundary markers: The instructions lack explicit delimiters or warnings to ignore potential instructions embedded within the PRDs or synthesis files.
- Capability inventory: The skill utilizes file system capabilities including reading, writing, creating directories, and copying files across the local workspace.
- Sanitization: No content validation or sanitization is performed on the data read from the PRD files before the agent acts upon it to update statuses or pipeline records.
Audit Metadata