trace

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly clones arbitrary remote repos via "--repo [url]" (ghq get -u "$URL") and queries GitHub issues/PRs with "gh issue list"/"gh pr list" in the SKILL.md workflow, meaning it ingests untrusted, user-generated web content (public repos/issues) that the agents read and use to influence search results and subsequent actions.