threads

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

• Third-party content exposure detected (high risk: 0.85). Outsider free text can enter the LLM context via the runtime unifapi calls to public Threads endpoints (threads/search/recent, threads/search/top, threads/search/profiles, threads/users/{username}/posts|replies|reposts), where the response includes outsider-authored text/biography content that the workflow then clusters and cites.