legal-search

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

• Third-party content exposure detected (high risk: 0.80). The SKILL.md explicitly instructs the agent to fetch and parse public third-party content (e.g., CourtListener API at https://www.courtlistener.com/api/rest/v4/, Harvard Case Law API at https://api.case.law/, and EUR-Lex endpoints) as part of its search workflow, so the agent would read and act on open web data that could contain untrusted inputs capable of influencing subsequent actions.