Inside Grammarly’s AI-driven automation with the MCP Server for Wiz

Grammarly outlines how adopting Wiz’s MCP server and LLM-enabled workflows (integrated via tools like Claude Desktop and LangGraph) automated tier-1 incident triage and related SOC tasks, reducing investigation time from 30–45 minutes to about four minutes (~90%) while expanding into threat hunting, detection engineering, and knowledge-driven intelligence. The report emphasizes an incremental approach—starting with high-impact, automatable use cases, iterating with real organizational data, and maintaining human-in-the-loop reviews—to scale enterprise security operations efficiently and consistently.