pm-review-board
Pass
Audited by Gen Agent Trust Hub on Jul 9, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill consists of instructional text and a reference checklist in markdown format. It does not include executable code, external dependencies, or instructions to perform network or system operations.
- [PROMPT_INJECTION]: The skill ingests untrusted user content (PRDs and prototypes) in SKILL.md. This constitutes an indirect prompt injection surface. However, the skill possesses no exploitable capabilities such as shell command execution, file system writes, or network requests (Capability inventory: text-only output). While explicit boundary markers and input sanitization are absent, the lack of powerful capabilities limits the potential impact of injection to the generated review text, resulting in a safe configuration.
Audit Metadata