recall

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

• Third-party content exposure detected (high risk: 0.70). The skill explicitly parses user-generated git commit messages and Pull Request/workflow data (via git log and gh CLI commands like "gh pr view" and "gh run list") as part of its mandated analysis in SKILL.md, which can contain untrusted instructions that influence classification and downstream decisions.