address-pr-comments

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly fetches and ingests GitHub PR review comments via the GitHub API (e.g., "gh api repos/{owner}/{repo}/pulls/{number}/comments" in SKILL.md Step 4 and references/gh-comment-api.md) and then reads and acts on those user-generated comment bodies (including suggestion blocks) to determine and apply code changes, so untrusted third-party content can directly influence actions.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The skill performs runtime calls to GitHub APIs (e.g., "gh api repos/{owner}/{repo}/pulls/{number}/comments --paginate" and "gh api graphql ...") to fetch AI reviewer comment bodies and suggestion blocks and then auto-applies those fetched suggestions as code edits and commits, meaning external content from those URLs directly controls code modifications at runtime.