pr-self-review

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

• Third-party content exposure detected (medium risk: 0.30). The skill’s required workflow reads outsider-authored PR content at runtime via GitHub API calls to fetch the PR body, discussion comments/reviews/issues-comments, and the whole PR diff (including any linked issue text referenced through those calls), so attacker-authored text can be ingested without prior item selection beyond the chosen PR target.