Last released Apr 8, 2026
Security monitor for AI coding agents — detect hook RCE, MCP poisoning, and API key theft before they execute
Supported by