gh-fix-ci

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

• Third-party content exposure detected (high risk: 0.85). Yes—scripts/inspect_pr_checks.py fetches outsider-authored PR review bodies/comments and CI log text via gh api/gh run view --log and then renders them as readable text (render_comprehensive_results / render_ci_results) that would be fed into the agent/LLM context for diagnosis and planning.