sn-search-social-en

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

• Third-party content exposure detected (high risk: 0.90). Yes — the skill programmatically fetches and returns user-generated content from public social platforms (reddit_search.py calls https://www.reddit.com/search.json, twitter_search.py calls the TikHub Twitter API, and youtube_search.py calls the YouTube Data API), so untrusted third‑party posts/tweets/videos are ingested and would be read/interpreted by the agent.