deep-research

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

• Third-party content exposure detected (high risk: 1.00). This skill's SKILL.md explicitly requires searching and collecting multiple external sources (see description "search, synthesize" and Workflow Phase 2 "Multi-Source Collection — Collect sources across different types") meaning the agent will fetch and ingest open/public third-party content as part of normal operation, which could include untrusted/user-generated material that influences its analysis and actions.