coderabbit-resolver

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

• Third-party content exposure detected (high risk: 1.00). The skill explicitly fetches and parses untrusted, user-generated GitHub PR review threads and issue comments (see workflows/review-loop.md Steps 1 & 6 and scripts/check-ci-status.sh / wait-for-ratelimit.sh / resolve-threads.sh) and uses those comment bodies to gate merges and drive actions, so third-party comments can materially influence the agent's behavior.