address-pr-reviews

Pass

Audited by Gen Agent Trust Hub on Apr 6, 2026

Risk Level: SAFE
Full Analysis
  • [PROMPT_INJECTION]: The skill is designed to handle untrusted input from PR comments securely.
  • Ingestion points: Processes PR review bodies and comments retrieved via the GitHub GraphQL API in SKILL.md.
  • Boundary markers: The skill explicitly classifies external feedback as "untrusted input" and establishes strict behavioral boundaries.
  • Capability inventory: Employs the GitHub CLI (gh) for API interactions and is authorized to modify repository files within the PR diff.
  • Sanitization: Restricts agent replies to fixed templates to prevent the echoing of potentially malicious instructions from comments.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 6, 2026, 01:51 PM