workflow-system

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

• Third-party content exposure detected (high risk: 0.90). The skill's Report Output workflow (reference/report-output.md) explicitly collects and parses git logs and PR data using commands like git log --oneline, gh pr diff, and gh pr view --json title,body,labels, then uses commit messages and PR bodies (user-generated/untrusted content) to generate changelogs and PR summaries and may run gh pr create, so untrusted third-party content is ingested and can materially influence agent decisions/actions.