x-scanner

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

• Third-party content exposure detected (high risk: 1.00). The skill (see SKILL.md and scripts/scan_x.py) calls xAI's Grok with the x_search tool to fetch real-time, public X/Twitter posts (untrusted, user-generated content) and ingests/summarizes those posts to produce flagged digests and drive follow-up actions (e.g., content ideas / Notion pushes), so third-party content can materially influence outputs and agent behavior.