Revit Converter Remote Mcp
About
The Revit Converter MCP is a remote MCP server that connects Claude (or any MCP-compatible AI client) directly to a Revit file conversion and BIM query service. Once configured, you can:
Details
- Author
- labsfuture
- Downloads
- 304
- Categories
- Other, AI
Jump to
- Upload Revit models to the server via URL.
- List all models on the server by name.
- Convert Revit models to XKT for 3D viewing.
- Query BIM data like element counts.
- Inspect element properties (e.g., random wall).
Setting up with Highlight
This MCP is not yet compatible with Highlight’s one-click setup. However, you can still use it with Highlight by following these steps:
- Download and install Highlight from highlightai.com/download
- Navigate to the plugins tab and select "Add Custom Plugin"
-
Configure the plugin with the settings below
Plugin Name
Revit Converter Remote McpCommand (node, npx, python, etc.)Please refer to the README for specific instructions on how to obtain API keys or other required environment variables.
- Enable "Start Automatically" if you want the plugin to start when Highlight launches
From the repository
Configure Claude Desktop by adding the provided JSON entry under mcpServers in the config file, replacing the Bearer token with your Labs Future API key (demo keys start with dk_, full keys with sk_). After restarting, Claude will have 7 tools available. Interact naturally by uploading a Revit model URL, converting it to XKT, or querying BIM data.
list_models
List all Revit models available on the server. Returns each model's name. Call this first to discover what's available before selecting or converting.
upload_model
Upload a Revit model to the server. Requires a demo (dk_) or full (sk_) API key — not available on the free public demo key. When the server is running remotely (e.g. Heroku), only fileUrl works — provide a public download link from Google Drive, Dropbox, WeTransfer, etc. Local filePath only works when the MCP server is running on the same machine as your files.
set_current_model
Select a model by name or keyword to make it the active model for this session.
convert_revit_to_xkt
Submit a Revit model for conversion to XKT format using Autodesk Design Automation. Free tier returns a sample model. Demo tier limited to 5 total conversions.
delete_model
Delete a Revit model you previously uploaded. You can only delete models you uploaded yourself — admin and other users' models cannot be deleted.
get_conversion_status
Check the status of a Revit-to-XKT conversion job. Call every 30-60 seconds until status is "complete" or "failed".
export_model
Export the active Revit model to IFC or DWG format via APS Model Derivative. Triggers the job and returns immediately — use get_export_status to poll progress and retrieve download links. Not available on free tier.
get_export_status
Check the status of a model export (IFC or DWG). When complete, returns direct download URLs valid for 60 minutes.
query_model
Ask a question about the active model's BIM data. Demo tier: 30 messages total. Free tier: 10 messages per session.
search_elements
Search for BIM elements in the active model by category and/or keyword.
Claude Desktop / Cursor
Paste into your MCP client config file to install this server.
{
"mcpServers": {
"revit converter remote mcp": {
"revit-converter": {
"command": "npx",
"args": [
"-y",
"mcp-remote",
"https://labsaps-d35810f00db3.herokuapp.com/mcp",
"--header",
"Authorization: Bearer labsfuture.comFreeDemo_MCPKey"
]
}
}
}
}McpServers
{
"revit-converter": {
"command": "npx",
"args": [
"-y",
"mcp-remote",
"https://labsaps-d35810f00db3.herokuapp.com/mcp",
"--header",
"Authorization: Bearer labsfuture.comFreeDemo_MCPKey"
]
}
}Getting Started
Step 1 — Get an API Key
Sign up at labsfuture.com and navigate to your account dashboard.
Generate Demo Key — free tier, includes 5 real Revit conversions to try the service
Generate API Key — for production / unlimited use
Get API Key
Demo keys start with dk_. Full API keys start with sk_.
Step 2 — Configure Claude Desktop
Open your Claude Desktop configuration file:
macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
Windows: %APPDATA%\Claude\claude_desktop_config.json
Add the following entry under mcpServers:
{
"mcpServers": {
"revit-converter": {
"command": "npx",
"args": [
"-y",
"mcp-remote",
"https://labsaps-d35810f00db3.herokuapp.com/mcp",
"--header",
"Authorization: Bearer labsfuture.comFreeDemo_MCPKey"
]
}
}
}
Replace the Bearer token with your own key from labsfuture.com. The key above is a shared demo key — get your own for personal model storage and full conversions.
MCP Config
Requires Node.js — npx must be available in your PATH. Install from nodejs.org if needed.
Step 3 — Restart Claude Desktop
After saving the config, restart Claude Desktop. Open Settings → Connectors and you should see revit-converter listed with 7 tools available.
Claude After Config
Usage Examples
Once connected, just talk to Claude naturally:
Upload a model
Could you upload this model to the revit converter server?
https://www.dropbox.com/scl/fi/yourfile.rvt?dl=0
Upload Model
List models on the server
Lists all models on the server by name
List Models
Convert to XKT and view in browser
yes, convert rac_basic_sample_project.rvt to xkt
Convert to XKT
Claude returns a direct viewer link — open it to see your model rendered in 3D:
View XKT
Query BIM data
how many walls in this model?
Query Walls
Inspect element properties
yes, just show me properties of a random wall
Element Data
About Labs Future
LabsFuture builds AI-native tools for the AEC industry. The Revit Converter MCP is part of a broader platform that brings BIM workflows into the age of conversational AI — letting architects, engineers, and developers interact with building data through natural language.
Visit www.labsfuture.com to:
Sign up and get your API key
Access the 3D web viewer
Learn about full platform features
Requirements
Claude Desktop (or any MCP-compatible client)
Node.js (for npx mcp-remote)
A Labs Future account with an API key
License
See LICENSE for details.
Sign in to leave a review
Use Google, GitHub, or an email account so ratings stay tied to real people.
No reviews posted yet.
