ci-fix

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

• Third-party content exposure detected (high risk: 0.80). The skill explicitly instructs the agent to fetch and inspect GitHub Actions run logs and artifacts using gh commands (e.g., gh run view --log, gh run view --log-failed, gh run download <id>), which pulls untrusted, user-generated content from public/third-party GitHub repos and PRs that the agent will read and interpret.