instagram-research

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

• Third-party content exposure detected (high risk: 0.85). The required runtime workflow collects public Instagram evidence (profiles/posts/comments/hashtags/search recall) from outsider-authored sources, which becomes readable text in the agent’s LLM context via the collection outputs (e.g., instagram-comments/instagram-posts/instagram-hashtags/instagram-search evidence items).