review

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly supports reviewing external PRs by URL and instructs the agent to run gh pr view / gh pr diff and to read the PR diff, files (including CLAUDE.md) as part of its workflow, which means it fetches and interprets untrusted, user-generated content from public GitHub repos that can materially influence tool use and decisions.