juicebox-hello-world

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

• Third-party content exposure detected (high risk: 0.70). The SKILL.md "Step 1: Create Search Script" shows the skill calling Juicebox's Search API (client.search.people) and printing returned profile data from that external service, which is untrusted/public user-generated content that the agent ingests and could influence subsequent actions.