xai-agent-tools

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

• Third-party content exposure detected (high risk: 0.90). This skill explicitly ingests public, untrusted third‑party content via the x_search tool (searching X/Twitter posts) and the web_search tool (real‑time web searches) — e.g., the Research and Financial agents that "Search X" and "Search the web" — which the agent reads and analyzes, enabling indirect prompt injection.