leaving-pr-comments
Pass
Audited by Gen Agent Trust Hub on Oct 27, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill defines how to process and format content for GitHub PRs, which is an external data surface.
- Ingestion points: Content prepared for GitHub comments, reviews, and replies (
SKILL.md). - Boundary markers: Commands the agent to prefix comments with a specific GitHub note callout to identify agent-authored content.
- Capability inventory: Text formatting and attribution for GitHub interactions.
- Sanitization: Includes an explicit instruction to "check the entire comment, including quoted text and attachments, for secrets and private details" before posting.
- [SAFE]: The skill focuses on transparency and attribution. It provides best-practice instructions for the agent to avoid leaking sensitive data.
Audit Metadata