mcp -iru-api

This MCP server provides AI-driven access to the Iru API (formerly Kandji API) through the Model Context Protocol. It exposes a suite of 23 MCP tools for device management, security monitoring, compliance reporting, user management, and configuration management, with smart caching, script generation, and structured response envelopes. You can interact with the server via Claude Desktop or other compatible MCP clients to perform tasks like searching and inspecting devices, evaluating security posture, generating compliant blueprints, and exporting large datasets with pagination-aware scripts. The server is designed to be multi-platform in the future, with planned integrations for Gemini CLI, Ollama, OpenAI ChatGPT, and Copilot Studio; Claude Desktop is production-ready today, while other platforms are documented in the ROADMAP.

To use it, configure your MCP client to point at the local or remote mcp-server-iru-api instance (as defined in mcp_config). Typical usage involves authenticating with your Kandji/Iru API token and subdomain via environment variables, starting the server, and then issuing natural language prompts or structured tool calls that map to the MCP tools (e.g., device search, vulnerability reports, compliance audits). The server returns standardized envelopes with metadata, results, and suggestions to guide your next actions.

Prerequisites

• Node.js 18 or higher installed on your machine or server
• Access to an Iru (Kandji) API token and subdomain
• Git installed for cloning the repository

Install from source

1. Clone the repository

   • git clone https://github.com/mangopudding/mcp-server-iru-api.git
   • cd mcp-server-iru-api

2. Install dependencies

   • npm install

3. Create an environment configuration

   • cp .env.example .env
   • Edit .env to include your Iru API credentials (example below)

4. Configure environment variables (example values)

   • KANDJI_API_TOKEN=your_api_token_here
   • KANDJI_SUBDOMAIN=your_subdomain_here
   • KANDJI_REGION=us

5. Verify API access (optional)

   • npm run verify

6. Run in development mode (auto-reload if supported by the project)

   • npm run dev

7. Build for production and start

   • npm run build
   • npm start

Notes and tips:

• Ensure your API token has the necessary permissions for the endpoints you plan to use.
• If you switch to a production deployment, consider using environment-specific configs and secure handling of tokens.
• The mcp_config example uses a placeholder path to the built server (dist/index.js). Replace /absolute/path/to/... with the actual path on your deployment target.
• For Claude Desktop or other MCP clients, ensure the client configuration references the correct server name (e.g., 'iru-api') as defined in mcp_config.
• If you encounter token or permission issues, run npm run verify or consult troubleshooting docs for endpoint permissions and scope.
• The server supports environment-variable-driven configuration; consider adding additional env vars as needed for advanced features or future platform integrations.