github-pr-workflow

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly fetches and parses user-generated content from public GitHub endpoints (e.g., curl calls to https://api.github.com/... for PRs, commits, and Actions logs shown in Sections 3–5 and references/ci-troubleshooting.md) and instructs the agent to read those logs/PR bodies and use them to decide and perform code fixes (the auto-fix loop), so untrusted third-party content can influence actions.