prd-reviewer
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill directs the agent to run a local Python script (plugins/testany-eng/scripts/trace_lint.py) to validate traceability metadata. This is a vendor-owned resource used for its intended purpose of linting document structure.
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process external PRD documents, creating an attack surface for indirect prompt injection.
- Ingestion points: The skill reads and parses PRD files and related project documentation provided by the user.
- Boundary markers: The instructions specify the use of markers (TRACEABILITY-METADATA:BEGIN/END) to define the scope of metadata analysis.
- Capability inventory: The agent is granted file system access to read documents and the ability to execute a shell command (python3).
- Sanitization: The skill employs a structured validation script for metadata validation, which acts as a sanitization layer for the ingested metadata block, though the script's internal logic is not provided for direct review.
Audit Metadata