revit
AI-Powered Revit Modeling
claude mcp add --transport stdio mcp-servers-for-revit-revit-mcp node <path to the built file>\build\index.js \ --env NODE_ENV="production" \ --env LOG_LEVEL="info"
How to use
revit-mcp exposes a set of MCP-powered tools that let AI clients drive and query a Revit project. After building the server, run it with Node so that MCP clients (such as Claude or other supported MCP clients) can connect and issue commands like get_current_view_info, create_point_based_element, or export_room_data. The server acts as the middleware between the MCP protocol and your Revit plugin setup, enabling AI-driven workflows to inspect models, create or modify elements, and run C# code within Revit via the plugin. To connect clients, configure the MCP client to point at the built server entry (for example, via the provided path to build/index.js). Restart the client to establish the MCP connection when the hammer icon indicates a healthy link. The server also supports a range of tools for querying and manipulating the model, including retrieving current view information, accessing elements, creating grids or rooms, and sending executable code to Revit through the plugin integration.
How to install
Prerequisites:
- Node.js 18+ installed on your machine
- npm available
- Access to revit-mcp-plugin and a Revit environment configured for the MCP workflow (as described in the related plugin repository)
Installation steps:
- Clone or download this repository and navigate to its folder.
- Install dependencies:
npm install
- Build the server:
npm run build
- Start the MCP server when ready (example):
node <path-to-built-file>/build/index.js
- Configure your MCP client (e.g., Claude) to connect to the server using the command/args pattern described in the client config (node + path to the built index.js). Restart the client and verify the connection indicator (hammer icon) shows a healthy link.
Additional notes
- The server is designed to be run with the accompanying revit-mcp-plugin on the Revit side; ensure the plugin is installed and configured per its documentation.
- Environment variables (e.g., NODE_ENV, LOG_LEVEL) can be used to adjust runtime behavior and logging.
- If you encounter path issues on Windows, ensure the path to the built file uses double backslashes or forward slashes as appropriate for your shell.
- This repo is being archived in favor of a mono-repo approach; rely on the npm package for installation and updates where possible.
- The MCP tools exposed include capabilities like get_current_view_info, get_available_family_types, create_point_based_element, delete_element, and send_code_to_revit, among others listed in the README.
Related MCP Servers
supermemory
Your memories are in ChatGPT... But nowhere else. Universal Memory MCP makes your memories available to every single LLM. No logins or paywall. One command to set it up.
obsidian -tools
Add Obsidian integrations like semantic search and custom Templater prompts to Claude or any MCP client.
Security-Detections
MCP to help Defenders Detection Engineer Harder and Smarter
mcp
Octopus Deploy Official MCP Server
furi
CLI & API for MCP management
bc-webclient
MCP server for Microsoft Dynamics 365 Business Central via reverse-engineered WebUI protocol. Enables AI assistants like Claude to interact with BC through the native WebSocket interface.