AEC Model Bridge

MCP server and native bridges for AI-assisted AEC workflows.

Available now

Coming soon

Future roadmap

Install | Tools | Documentation | Latest release

AEC Model Bridge connects MCP clients such as Claude Desktop, VS Code, GitHub Copilot, and custom agents to Revit and a growing ecosystem of AEC platforms.

The Python hub handles MCP communication and routes each request to the relevant desktop, headless, cloud, or compute provider. Revit commands run through a native C# add-in and ExternalEvent, keeping API work on Revit's main thread.

Platform Status

The status below reflects the Omni-Bridge system blueprint as of June 12, 2026. Preview integrations are present but still have incomplete workflows or hardening tasks.

Available Now

Revit, IfcOpenShell, and Speckle are available integrations. Navisworks currently exposes bridge health and document metadata; model-tree, append/refresh, viewpoints, and clash workflows are next. Rhino.Compute is integrated, while live Rhino MCP connectivity remains a preview.

Coming Soon

This wave adds Excel and SharePoint workbook round trips, ACC/Forma workflows through Autodesk Platform Services, Solibri model checking, and Power BI templates backed by Speckle or a local Parquet/DuckDB data plane.

Future Roadmap

Future connectors are opportunity-driven and will be scheduled when each has a defined workflow, test access, and a supported API path.

Highlights

• 100 MCP tools for model authoring, documentation, parameters, views, sheets, exports, worksharing, architecture, structure, MEP, geometry, and QA.
• Native Revit add-in with no pyRevit or Dynamo dependency.
• Revit 2024, 2025, 2026, and 2027 support.
• Headless IFC, Speckle cloud, Rhino.Compute, and cross-platform provider architecture.
• Reflection and in-process Python for advanced API workflows.
• Localhost-only bridge by default.
• Mock mode for development and automated testing without Revit.

How It Works

MCP client
    |
    | MCP over stdio
    v
Python MCP server
    |
    | HTTP on 127.0.0.1:3000
    v
AEC Model Bridge add-in
    |
    | ExternalEvent
    v
Revit API

Revit API Access

The typed MCP tools cover the common workflow surface and are the recommended default for agents.

For work outside the tool catalog, invoke_method, reflect_get, and reflect_set can work with public .NET API members. execute_python runs IronPython inside Revit with doc, uidoc, uiapp, and app available.

These advanced tools run with the same permissions as the Revit process. Keep the bridge on localhost and only use trusted MCP clients and prompts.

Development

# Python tests
python -m pytest packages/mcp-server-revit/tests

# Build one Revit version
$RevitVersion = Read-Host "Revit year (2024, 2025, 2026, or 2027)"
.\scripts\build-addin.ps1 -RevitVersion $RevitVersion -Configuration Release

# Build all supported versions
.\scripts\package.ps1 -RevitVersion All

CI builds the Python server and add-in targets for Revit 2024 through 2027.

Documentation

• Installation guide
• Tool reference
• Architecture
• Multi-platform integration handover
• Configuration reference
• Security
• MCP clients and registry
• Contributing

Project

Maintained by A. Sam Mohammad. LinkedIn | Issues

Version 1.1.0 and later is available under your choice of GPL-3.0-or-later with the Revit Linking Exception, or a separate commercial license. The GPL option permits community use while allowing the add-in to operate through Autodesk Revit APIs. Commercial terms are available for proprietary distribution and negotiated requirements. Version 1.0.2 and earlier remains available under the MIT License.

AEC Model Bridge is an independent project and is not sponsored, endorsed, or provided by Autodesk. Autodesk and Revit are trademarks of the Autodesk group of companies. See TRADEMARKS.md.

Installation

Requirements

You also need Windows 10 or 11, Python 3.11 or later, and a licensed Revit installation for the version you want to use.

1. Install the MCP server

git clone https://github.com/Sam-AEC/aec-model-bridge.git
cd aec-model-bridge

py -m venv .venv
.\.venv\Scripts\Activate.ps1
python -m pip install -e packages/mcp-server-revit

2. Install the Revit add-in

Set the version to match your Revit installation:

$RevitVersion = Read-Host "Revit year (2024, 2025, 2026, or 2027)"

.\scripts\package.ps1 -RevitVersion $RevitVersion
.\scripts\install.ps1 -RevitVersion $RevitVersion

The installer places version-specific binaries in:

C:\ProgramData\AECModelBridge\bin\<year>

The add-in manifest is installed per user in:

%APPDATA%\Autodesk\Revit\Addins\<year>

Use -AllUsers with install.ps1 to install the manifest under C:\ProgramData\Autodesk\Revit\Addins\<year> instead.

To prepare binaries for every supported version in one pass:

.\scripts\package.ps1 -RevitVersion All

Prebuilt packages are attached to GitHub releases when available. The source installation above is the canonical path for all supported Revit versions.

3. Configure your MCP client

Use the Python executable from the virtual environment and choose a workspace that the server may access:

{
  "mcpServers": {
    "aec-model-bridge": {
      "command": "C:\\path\\to\\aec-model-bridge\\.venv\\Scripts\\python.exe",
      "args": ["-m", "revit_mcp_server.mcp_server"],
      "env": {
        "MCP_REVIT_MODE": "bridge",
        "MCP_REVIT_BRIDGE_URL": "http://127.0.0.1:3000",
        "MCP_REVIT_WORKSPACE_DIR": "C:\\RevitProjects",
        "MCP_REVIT_ALLOWED_DIRECTORIES": "C:\\RevitProjects"
      }
    }
  }
}

VS Code users can start from .vscode/mcp.json.

Clients that support MCP Bundles can install the .mcpb file from the latest release. The Revit add-in is still required because the MCP server communicates with the running desktop application.

4. Verify the connection

Restart Revit after installing the add-in, open a model, and run:

Invoke-RestMethod http://127.0.0.1:3000/health

The response should report healthy and the active Revit version.