hunter-22

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

• Third-party content exposure detected (medium risk: 0.30). SKILL.md requires calling ClawHunter’s free discovery endpoint (POST https://clawhunter.fun/api/v1/match) and then inspecting each returned candidate’s body/url (including for GitHub repo links), which means the workflow ingests outsider-authored bounty text from the indexed marketplace without first selecting a specific item.