pr-review-canvas

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

• Third-party content exposure detected (medium risk: 0.30). The skill’s required runtime workflow uses gh api to read outsider-authored GitHub PR content (PR title/body, file patches/diffs, and PR comments) via repos/{owner}/{repo}/pulls/{number} / .../files / .../comments, then has the LLM analyze and write an HTML review from those texts.