leaving-pr-comments
Pass
Audited by Gen Agent Trust Hub on Oct 28, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from external sources (GitHub pull request descriptions, comments, and review threads) to generate responses on behalf of the user.
- Ingestion points: Ingests PR content, review replies, and collaborator feedback from GitHub to construct responses (defined in
SKILL.md). - Boundary markers: Implements a mandatory attribution block (
> [!NOTE]) to identify and separate agent-authored content from the rest of the conversation. - Capability inventory: The skill is designed to manage GitHub write operations, such as posting comments and resolving reviews, and utilizes the
ghCLI tool for metadata retrieval (SKILL.md). - Sanitization: Explicitly instructs the agent to perform a safety check on the generated comment and any attachments for secrets or private details before posting (
SKILL.md).
Audit Metadata