HOL Guard

Overview

HOL Guard is an MCP server that brings local-first AI agent security evidence and approval workflows directly into your development environment. By implementing the Model Context Protocol, it lets AI assistants call security tools natively rather than requiring manual context switching between applications.

The server runs entirely locally as the Python package hol-guard, installed via uvx. It communicates over stdio with no authentication required, making it ready to use immediately after installation. Version v3.0.31 is available under the Apache-2.0 license, with source code maintained at hashgraph-online/hol-guard on GitHub, where the project has accumulated 508 stars.

HOL Guard appears in the AI & Agents category on mcp.site and works with all major MCP clients including Claude Desktop, Claude Code, Cursor, VS Code, Windsurf, and Zed. If you maintain this project, you can claim this listing to verify ownership and ensure the information stays up to date.

Installation

// claude_desktop_config.json
{
  "mcpServers": {
    "hol_guard": {
      "command": "uvx",
      "args": [
        "hol-guard"
      ]
    }
  }
}