LeanIX MCP Integration

What is LeanIX MCP Integration?

A Model Context Protocol (MCP) server that connects LeanIX to AI assistants. It exposes LeanIX’s GraphQL API as five MCP tools, enabling AI assistants to read and write LeanIX fact sheets. It is designed for developers who want to interact with their LeanIX workspace through conversational interfaces like Claude.

How to use LeanIX MCP Integration?

Ensure Node.js v14+ and a LeanIX workspace with an API token. Clone the repository, run npm install, and create a .env file with LEANIX_SUBDOMAIN and LEANIX_TOKEN. For Claude Desktop, add a server configuration to its config file pointing to server.js. Start the server with node server.js and test through any MCP‑compatible client.

Key features of LeanIX MCP Integration

• Fact Sheet Overview: get counts and statistics of fact sheets
• Search: find fact sheets by name
• Subscription Management: view subscribers of specific fact sheets
• Create Fact Sheets: add new fact sheets to the workspace
• Update Fact Sheets: modify existing fact sheet information

Use cases of LeanIX MCP Integration

• Query workspace‑level fact sheet statistics through an AI assistant
• Search for specific fact sheets without opening the LeanIX UI
• Manage subscriptions and understand who follows a fact sheet
• Automate creation and updates of fact sheets from chat prompts

FAQ from LeanIX MCP Integration

What dependencies or runtime does LeanIX MCP Integration require?

Node.js v14 or higher, a LeanIX workspace with an API token, and basic familiarity with GraphQL and MCP.

Where does data live and how is authentication handled?

All data remains in your LeanIX workspace. Authentication uses a LeanIX API token stored in the .env file as LEANIX_TOKEN, along with the workspace subdomain.

What transport or protocol does this server use?

The server communicates via MCP over stdio (Node.js process) and works with any MCP‑compatible client, such as Claude Desktop.

How are GraphQL schema changes handled?

Always verify the current LeanIX GraphQL schema in their documentation or GraphiQL interface, as it may change over time.

What is the required response format for MCP tools?

All tool responses must be an object with a content array containing objects with type: "text" and a text field.