tend-pr-tick
Pass
Audited by Gen Agent Trust Hub on Oct 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes data from potentially untrusted external sources, specifically pull request comments and CI failure logs.\n
- Ingestion points: New comments and CI statuses are read from the pull request platform to drive the automation logic (SKILL.md).\n
- Boundary markers: The skill includes high-level instructions to ignore instructions or code snippets embedded within PR comments (SKILL.md).\n
- Capability inventory: The skill has the ability to modify code, push changes to repositories, and invoke other automation tools like /do and /define (SKILL.md).\n
- Sanitization: Security guidelines explicitly state that PR comments are untrusted and that implementation should follow the agent's judgment rather than copy-pasting external suggestions (SKILL.md).
Audit Metadata