grok-search
Warn
Audited by Snyk on Feb 14, 2026
Risk Level: MEDIUM
Full Analysis
MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).
- Third-party content exposure detected (high risk: 0.90). This skill invokes xAI Grok's web_search and x_search (see scripts/grok_search.mjs and the SKILL.md --web / --x commands) to fetch web pages and X/Twitter posts (public, user-generated content) which are returned as JSON and intended to be read/processed by the agent, exposing it to untrusted third-party content that could carry indirect prompt injection.
Audit Metadata