WHAT IT CONNECTS
What ZedMoster revit mcp gives an AI Agent
The source catalog does not include a structured feature list. Review the repository documentation before use.
SECURITY
An MCP listing is not a security review
An MCP server may receive model context, credentials, local files, or permission to call external systems. Review its code, requested environment variables, network behavior, package provenance, and maintenance status before connecting it to an agent.
README
Repository documentation
xml.Revit.MCP provides a comprehensive set of tools for integrating with Autodesk Revit through the Model Context Protocol (MCP). This library serves as a bridge between AI assistants and Revit, enabling powerful automation capabilities and programmatic interaction with building models.
The tool library includes numerous functions for Revit automation and interaction:
Basic Operations:
- Get available commands from Revit plugin
- Execute specified commands in Revit
- Call specific Revit functions with parameters
- Retrieve view data and selected elements
Element Management:
- Find elements by category
- Get element parameters and locations
- Update element parameters
- Delete elements
- Show/highlight elements in current view
- Move elements to new positions
Creation Tools:
- Create levels/floors
- Create floor plan views
- Create grid lines
- Create walls and floors
- Create rooms and room tags
- Create doors and windows
- Create MEP elements (ducts, pipes, cable trays)
- Create family instances
- Link DWG files
- Create sheets
- xml.Revit: Version 1.3.4.3 or newer
- Python: 3.10 or newer
- UV Package Manager: Required for installation
- Revit: Compatible with versions 2019-2024 (with plugin)
-
First, install the UV package manager:
pip install uv
-
Install the revit-mcp package:
pip install revit-mcp
-
Test the installation:
uvx revit-mcp
You should see:
RevitMCPServer - INFO - Successfully connected to Revit on startup
Edit claude_desktop_config.json to include:
Edit mcp.json to include:
Edit cline_mcp_setting.json to include:
{
"mcpServers": {
"RevitMCPServer": {
"disabled": false,
"timeout": 30,
"command": "uvx",
"args": ["revit-mcp"],
"transportType": "stdio",
"autoApprove": [
"active_view",
"call_func",
"create_cable_trays",
"create_door_windows",
"create_ducts",
"create_family_instances",
"create_floors",
"create_floor_plan_views",
"create_grids",
"create_levels",
"create_pipes",
"create_room_separation_lines",
"create_room_tags",
"create_rooms",
"create_sheets",
"create_walls",
"delete_elements",
"execute_commands",
"find_elements",
"get_commands",
"get_locations",
"get_selected_elements",
"get_view_data",
"link_dwg_and_activate_view",
"move_elements",
"parameter_elements",
"show_elements",
"update_elements"
]
}
}
}You can create custom MCP DLL files to implement additional functionality by:
- Implementing the
xml.Revit.MCP.Public.IMCPMethodinterface - Following JSON-RPC 2.0 specification for communication
- Compiling to a DLL and placing it in the designated MCP folder
When using the revit-mcp-plugin:
- Register the plugin with Revit
- Configure commands through: Add-in Modules → Revit MCP Plugin → Settings
- Enable the service: Add-in → Revit MCP Plugin → Revit MCP Switch
Once enabled, AI assistants can discover and control your Revit program, executing the various commands provided by the xml.Revit.MCP tools library.
COMMUNITY
