code-review
Pass
Audited by Gen Agent Trust Hub on Oct 4, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No security issues were detected. The skill instructions are focused on providing high-quality code review feedback and explicitly command the agent to honor existing repository safety policies and refrain from unauthorized write actions.\n- [INDIRECT_PROMPT_INJECTION]: The skill inherently processes external, untrusted code and design documents, which creates a surface for indirect prompt injection. The risk is mitigated by the skill's specific focus and operational constraints.\n
- Ingestion points: Code plan drafts, specs, diffs, and implementation shapes as described in
SKILL.md.\n - Boundary markers: No specific delimiters or boundary markers are instructed for the input data.\n
- Capability inventory: Instructions explicitly prohibit unauthorized edits, commits, or pushes.\n
- Sanitization: No sanitization or filtering of external content is specified.
Audit Metadata