fix-pr

CRITICAL E004: Prompt injection detected in skill instructions.

• Potential prompt injection detected (medium risk: 0.30). The skill contains detailed PR-fixing workflow instructions that explicitly require stripping or denying author-identifying commit trailers (e.g., Co-authored-by, Signed-off-by, "Made with Cursor") and applying "trailer hygiene," which weakens provenance/attribution and is suspicious steering of metadata handling.

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (medium risk: 0.30). The skill reads outsider-authored GitHub pull-request review thread bodies, review comments, and issue/PR conversation comments at runtime via its required hunt steps (GraphQL pullRequest.reviewThreads + comments, and REST pulls/<number>/comments, pulls/<number>/reviews, and issues/<number>/comments).