leaving-pr-comments
Pass
Audited by Gen Agent Trust Hub on Nov 4, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes content intended for GitHub PR comments, including quoted text and attachments, which represents an input surface for untrusted data that could potentially contain malicious instructions.
- Ingestion points: Content of Pull Request comments, quoted text, and attachments (SKILL.md).
- Boundary markers: The skill defines a specific header format for attribution but does not provide explicit delimiters for separating untrusted external data within the comment body.
- Capability inventory: The skill utilizes GitHub API tools to post comments and resolve review threads.
- Sanitization: The skill includes a specific instruction to check the entire comment for secrets and private details before posting.
Audit Metadata