gh-fix-ci

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

• Third-party content exposure detected (high risk: 0.85). The required workflow runs scripts/inspect_pr_checks.py, which uses gh pr checks and gh run view --log / gh api .../actions/jobs/.../logs to fetch GitHub Actions log text from the PR’s failing checks (outsider-authored CI output) and then includes extracted “Failure snippet” text in the agent’s context.

MEDIUM W013: Attempt to modify system services in skill instructions.

• Attempt to modify system services in skill instructions detected (high risk: 0.80). The skill explicitly instructs rerunning commands with sandbox_permissions=require_escalated and to use escalated gh scopes/keyring access, which asks the agent to bypass sandboxing or escalate its permissions (a security-sensitive action).