ci-debug

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 parses GitHub Actions job logs and PR/run data from external repositories using gh api/gh pr checks (see the "Fetch the failing log" and "Resolve the failing job" execution steps), so untrusted third-party log content is read and used to classify failures and propose fixes.