ship-safety
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted code diffs, which provides a surface for indirect prompt injection. A malicious diff could contain instructions aimed at tricking the agent into approving unsafe code or misreporting findings. However, this is inherent to the skill's primary purpose of code review.
- Ingestion points: The skill operates on "diffs" provided as input, as specified in the SKILL.md body.
- Boundary markers: The instructions do not define specific delimiters to separate the diff content from the analysis instructions.
- Capability inventory: The skill is restricted to generating textual analysis and does not have file-system, network, or command execution capabilities.
- Sanitization: There is no instruction to sanitize or validate the content of the diff before analysis.
Audit Metadata