github

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

• Third-party content exposure detected (high risk: 0.80). This skill explicitly instructs the agent to fetch and read user-generated content from public GitHub repositories (e.g., via gh pr checks, gh run view --log-failed, and gh api), so untrusted third-party PR descriptions, comments, and CI logs could be ingested and influence follow-up actions.