Skip to main content
Glama

JumpCloud MCP Server

by hoium

🤖 JumpCloud MCP Server

A natural language API server and agent for your JumpCloud environment, built with FastAPI. Supports the Model Context Protocol (MCP) for integration with AI assistants and code editors.

This MCP server lets you:

  • 🔎 Query users, systems, groups, and SSO apps via REST

  • 💬 Ask natural language questions via /ask

  • 🤖 Use a local, LLM-free agent (keyword-based tool matcher)

  • 🐳 Run everything in Docker

  • ⚙️ Integrate with MCP-compatible clients (Claude Desktop, Cursor, etc.)


📦 Features

  • ✅ FastAPI-based REST API for JumpCloud data

  • 🔐 Token authentication using x-api-key

  • 🤖 /ask endpoint for semantic/natural language queries

  • 🐳 Docker Support

  • 💡 MCP protocol support for AI assistants and code editors


🛠️ Quick Setup

1. Clone and configure environment

git clone https://gitlab.com/barkada/itops/jumpcloud-mcp cd jumpcloud-mcp cp .env.example .env

Update .env with your keys:

JUMPCLOUD_API_KEY=your_jumpcloud_api_key MCP_API_URL=http://localhost:8000

2. Build and run with Docker

docker-compose up --build

The server will start on http://localhost:8000.


3. Call MCP via REST

curl -X GET http://localhost:8000/systems -H "x-api-key: $JUMPCLOUD_API_KEY"

4. Ask with natural language

curl -X POST http://localhost:8000/ask -H "Content-Type: application/json" -H "x-api-key: $JUMPCLOUD_API_KEY" -d '{"prompt": "List all active Mac systems"}'

📁 Directory Structure

jumpcloud_mcp/ ├── main.py # FastAPI app + MCP protocol + /ask endpoint ├── jumpcloud/ │ ├── client.py # JumpCloud API calls: users, systems, groups │ ├── models.py # Pydantic models for validation │ ├── mcp_agent_runner.py # Keyword-based tool-matching agent (no LLM) │ └── auth.py # API key auth ├── .env # Secrets/config ├── Dockerfile # Build FastAPI server container ├── docker-compose.yml # Docker Compose for dev/prod ├── requirements.txt # Python dependencies (NO openai/anthropic) └── README.md # Docs and usage guide

🔧 REST API Reference - API Docs

📍 GET Endpoints

  • /users

  • /systems

  • /user-groups

  • /system-groups

  • /sso-applications

📍 POST

  • /ask — Accepts {"prompt": "..."}

  • /users/search Search JumpCloud users using filters and fields.

    • {"filter": [{"department": "IT"}], "fields": "email username sudo"}

  • /commands/search Search JumpCloud commands using filters and fields.

    • {"filter": [{"command": "restart"}], "fields": "name command sudo"}


💡 MCP Client Integration

This server supports the Model Context Protocol (MCP) and can be used with various AI assistants and code editors.

Claude Desktop

Add to your Claude Desktop configuration (~/Library/Application Support/Claude/claude_desktop_config.json on macOS):

{ "mcpServers": { "jumpcloud-mcp": { "command": "uvicorn", "args": ["main:app", "--host", "0.0.0.0", "--port", "8000"], "cwd": "/path/to/jumpcloud_mcp" } } }

Cursor IDE

Create .cursor/mcp.json in your workspace:

{ "mcpServers": { "jumpcloud-mcp": { "url": "http://localhost:8000", "description": "JumpCloud MCP Server" } } }

Other MCP Clients

For any MCP-compatible client, configure it to connect to:

  • HTTP URL: http://localhost:8000

  • Protocol: MCP over HTTP

  • Authentication: Include x-api-key header with your JumpCloud API key


✨ Support

This project is maintained for local/private JumpCloud automation and is ideal for secure deployments, development, and custom integrations with MCP-compatible AI assistants and code editors.


-
security - not tested
F
license - not found
-
quality - not tested

hybrid server

The server is able to function both locally and remotely, depending on the configuration or use case.

Enables natural language interaction with JumpCloud environments to query users, systems, groups, and SSO applications. Features a local LLM-free agent for keyword-based tool matching and REST API access to JumpCloud data.

  1. 📦 Features
    1. 🛠️ Quick Setup
      1. 1. Clone and configure environment
      2. 2. Build and run with Docker
      3. 3. Call MCP via REST
      4. 4. Ask with natural language
    2. 📁 Directory Structure
      1. 🔧 REST API Reference - API Docs
        1. 📍 GET Endpoints
        2. 📍 POST
      2. 💡 MCP Client Integration
        1. Claude Desktop
        2. Cursor IDE
        3. Other MCP Clients
      3. ✨ Support

        MCP directory API

        We provide all the information about MCP servers via our MCP API.

        curl -X GET 'https://glama.ai/api/mcp/v1/servers/hoium/jumpcloud_mcp'

        If you have feedback or need assistance with the MCP directory API, please join our Discord server