Aguara MCP

Aguara MCP is a local security engine for autonomous AI agents integrating with the Model Context Protocol (MCP). It provides AI agents (like Claude Code, Cursor, and others) with tools to detect and block security threats—such as prompt injection, credential leaks, command execution, data exfiltration, and supply-chain attacks—when reviewing third-party content, agent skills, plugin manifests, or tool configurations. Aguara MCP runs entirely locally as a Go library, ensuring no network access or LLM calls, and provides structured reports for agent decision-making before actions are taken.