process-prs
Pass
Audited by Gen Agent Trust Hub on Sep 1, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from pull requests, creating a surface for indirect prompt injection. Ingestion points: gh pr view and git diff in SKILL.md. Boundary markers: Absent. Capability inventory: gh commands and /verify skill in SKILL.md. Sanitization: Absent. Risk is mitigated by tiering that restricts execution to trusted members.
- [DYNAMIC_EXECUTION]: The skill invokes /verify to run code during PR validation. This is restricted to PRs from repository members, limiting the exposure to untrusted code.
- [SAFE]: No malicious patterns found. The skill enforces least privilege and author-based risk segmentation.
Audit Metadata