omh-deep-research

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

• Third-party content exposure detected (high risk: 0.90). This skill explicitly requires the web toolset and dispatches [omh-role:researcher] subagents in Phase 2 that run web_search/web_extract against open/public URLs, and the parent then inlines those findings into the synthesist/verifier contexts in Phases 4–5—meaning untrusted third-party content is fetched and directly used to drive synthesis, verification, and state-changing actions.