The MCP Server for Revit integration provides an AI-powered control layer that enables model context management and protocol-driven interactions within Revit workflows. It encapsulates the Model Context Protocol (MCP) to streamline communication between tools and the Revit environment.

🛠️ Key Features

AI-powered control for Revit data and actions
Model Context Protocol (MCP) integration
Lightweight server suitable for automation pipelines

🚀 Use Cases

Revit automation tasks with context-aware commands
Integrating external tools via MCP for Revit sessions
Scripting and orchestration of Revit workflows

⚡ Developer Benefits

Clear, protocol-based interaction model
Focused server for Revit-centric automation
Facilitates integration with AI tooling and pipelines

⚠️ Limitations

Description provided limited; no detailed specs or README excerpt available
Specific MCP version, endpoints, and data schemas not disclosed in source data

Revit MCP - AI-Powered Revit Control

English | 繁體中文

Revit MCP lets AI clients call Autodesk Revit tools through the Model Context Protocol (MCP). The MCP server forwards tool calls to a local Revit add-in, and the add-in executes the corresponding Revit API workflow.

Demo video: Revit MCP - AI-Powered BIM Workflow Demonstration
Knowledge site: https://shuotao.github.io/REVIT_MCP_study/
Default WebSocket port: 8964

What is this?

Talk to Revit in plain language. Ask your AI client to "dimension every wall on this view" or "check the curtain-wall elevations", and Revit does it — through 166 MCP tools backed by 72 professional BIM SOPs (building code, quantity take-off, compliance checks).

Who it's for: BIM engineers and architects who use Revit and want AI-assisted, standards-based workflows. You'll need Revit (2022–2026) on Windows and to be comfortable installing an add-in.

Quickstart (3 steps)

Install the Revit add-in. Build and deploy the C# add-in — see Manual Setup. This is the half that actually talks to Revit.

Point your AI client at the MCP server. No cloning needed — it runs straight from npm:

json

{ "mcpServers": { "revit-mcp": { "command": "npx", "args": ["-y", "@shuotao/revit-mcp-server"] } } }

Open Revit, enable the MCP service in the ribbon, and start asking. Full setup: AI Client Configuration.

Questions or want to show what you built? → Discussions

Current Project Counts

Item	Count	Source
Runtime MCP tools	166	`registerRevitTools()` in `MCP-Server/src/tools/index.ts`
Domain SOP files	72	`domain/.md` except `README.md`, plus `domain/references/.md`
Claude skills	50	`.claude/skills/*/SKILL.md`

When these numbers change, update CLAUDE.md, README.zh-TW.md, this file, docs/DOCUMENT_AUDIENCE_INVENTORY.md, and run:

powershell

.\scripts\verify-qaqc.ps1 -SkipBuild -SkipDeploy

Architecture

text

AI Client
  Claude Desktop / Claude Code / Gemini CLI / VS Code Copilot / Antigravity
        |
        | stdio
        v
MCP Server
  Node.js / TypeScript
  MCP-Server/build/index.js
        |
        | WebSocket ws://localhost:8964
        v
Revit Add-in
  C# / Revit API
  MCP/Application.cs
  MCP/Core/SocketService.cs
  MCP/Core/ExternalEventManager.cs
        |
        v
Autodesk Revit

External AI clients do not need an API key inside this repository. Their account and authorization are managed by the AI client itself. Only an embedded Revit chat feature that directly calls an AI API would need an API key.

Requirements

Item	Requirement
OS	Windows 10 or later
Revit	Autodesk Revit 2022, 2023, 2024, 2025, 2026
.NET	.NET Framework 4.8 for Revit 2022-2024; .NET 8 for Revit 2025-2026
Node.js	LTS, preferably 20.x or later

One-Click Setup

Recommended for new users:

powershell

.\scripts\setup.ps1

For AI agents or non-interactive setup:

powershell

powershell -ExecutionPolicy Bypass -File scripts/setup.ps1 -NonInteractive -RevitVersions "2024,2025"

The setup script checks prerequisites, installs dependencies, builds the MCP server, builds and deploys the Revit add-in, and helps configure common AI clients.

Install from MCP Registry

The MCP server is published on the MCP Registry as io.github.shuotao/revit-mcp-server (npm package @shuotao/revit-mcp-server). Run it directly with:

bash

npx -y @shuotao/revit-mcp-server

Note: this npm package is only the Node stdio bridge. The C# Revit add-in (MCP/) must still be installed separately — follow Manual Setup below for your Revit version.

Manual Setup

1. Build the MCP Server

powershell

cd MCP-Server
npm install
npm run build

AI clients launch:

text

node MCP-Server/build/index.js

2. Build the Revit Add-in

Choose the configuration that matches your Revit version:

powershell

cd MCP
dotnet build -c Release.R22 RevitMCP.csproj   # Revit 2022
dotnet build -c Release.R23 RevitMCP.csproj   # Revit 2023
dotnet build -c Release.R24 RevitMCP.csproj   # Revit 2024
dotnet build -c Release.R25 RevitMCP.csproj   # Revit 2025
dotnet build -c Release.R26 RevitMCP.csproj   # Revit 2026

Expected output:

text

MCP/bin/Release.R{YY}/RevitMCP.dll

Example for Revit 2024:

text

MCP/bin/Release.R24/RevitMCP.dll

3. Deploy the Add-in

Recommended:

powershell

.\scripts\install-addon.ps1

For manual deployment, place the .addin file and DLL under the matching Revit Addins directory, and keep the relative assembly path in RevitMCP.addin:

xml

<Assembly>RevitMCP\RevitMCP.dll</Assembly>

Do not create version-specific .addin files, and do not hardcode absolute DLL paths.

AI Client Configuration

Project-level .mcp.json:

json

{
  "mcpServers": {
    "revit-mcp": {
      "type": "stdio",
      "command": "node",
      "args": ["./MCP-Server/build/index.js"],
      "env": {}
    }
  }
}

VS Code config in .vscode/mcp.json:

json

{
  "servers": {
    "revit-mcp": {
      "type": "stdio",
      "command": "node",
      "args": ["${workspaceFolder}/MCP-Server/build/index.js"],
      "env": {}
    }
  }
}

Other AI clients use the same concept: launch MCP-Server/build/index.js with node.

Config template per client:

AI Client	Config location	Template
Claude Code	project root `.mcp.json`	built in, works out of the box
Claude Desktop	`%APPDATA%\Claude\claude_desktop_config.json`	`MCP-Server/claude_desktop_config.json`
Gemini CLI	`~/.gemini/settings.json`	`MCP-Server/gemini_mcp_config.json`
VS Code Copilot	`.vscode/mcp.json`	built in
Antigravity	UI settings	`Antigravity_MCP_Complete_Guide.md`

Replace <YOUR_PROJECT_PATH> in the templates with the actual project path on your machine.

Switching Between AI Clients

The Revit-side WebSocket service accepts only one MCP connection at a time: a newly connected MCP server replaces the previous connection. Multiple AI clients are therefore used by switching, not concurrently:

Close the current AI client (or disable its MCP server).
Start the other AI client; once its MCP server connects to localhost:8964, it takes over.
If the connection misbehaves, restart the MCP service from the Revit ribbon to reset it.

Startup Flow

Start Revit.
Open or create a Revit project.
Enable the MCP service from the Revit ribbon.
Confirm the Revit add-in is listening on localhost:8964.
Start or restart the AI client so it loads the MCP server.
Call Revit MCP tools from the AI client.

If localhost:8964 is unreachable, Revit may not be running, the MCP service may be off, the port may be occupied, or the AI client and Revit add-in may be using different port settings.

Project Structure

text

REVIT_MCP/
  MCP/                         Revit Add-in (C#)
    Application.cs             Revit add-in entry point
    RevitMCP.csproj            Single multi-version project
    RevitMCP.addin             Single add-in manifest
    Core/
      SocketService.cs         Revit-side WebSocket server
      ExternalEventManager.cs  UI-thread execution bridge
      RevitCompatibility.cs    Revit 2022-2026 compatibility helpers
      CommandExecutor.cs       Main command dispatcher
      Commands/*.cs            Command modules
  MCP-Server/                  MCP Server (Node.js / TypeScript)
    src/index.ts               stdio MCP server entry
    src/socket.ts              WebSocket client to Revit
    src/tools/*.ts             MCP tool definitions
  domain/                      Shared BIM SOPs; do not convert to English-only
  .claude/                     AI commands and skills
  docs/                        Human-facing docs and public knowledge site
  scripts/                     Setup, deployment, QA/QC scripts
  log/                         Append-only session and commit logs

AI Docs and Human Docs

Type	Location	Rule
AI-only	`CLAUDE.md`, `.claude/commands/`, `.claude/skills/`	English-first to avoid mojibake
Human-facing	`README.md`, `README.zh-TW.md`, `docs/`, `scripts/README.md`	Match the reader's language
Shared	`domain/*.md`, `log/README.md`	Domain files must remain Chinese-readable and must not become English-only
Historical	`docs/_archive/**`, old logs	Preserve by default

See docs/DOCUMENT_AUDIENCE_INVENTORY.md.

Domain, Skill, and Tool Responsibilities

domain/*.md: BIM SOPs, regulatory logic, and calculation methods. Shared by humans and AI.
.claude/skills/*/SKILL.md: AI workflow orchestration.
MCP-Server/src/tools/*.ts: MCP tool definitions and input schemas.
MCP/Core/Commands/*.cs: Revit API implementation.

If a Domain file and a Skill disagree on method, the Domain file wins.

QA/QC

After documentation, tool, Domain, Skill, build, or deployment changes, run:

powershell

.\scripts\verify-qaqc.ps1 -SkipBuild -SkipDeploy

Before deployment, run a full check:

powershell

.\scripts\verify-qaqc.ps1 -Version 2024

QA/QC checks:

forbidden legacy files and paths
required file structure
README / CLAUDE / docs count alignment
Domain table forward and reverse coverage
local Markdown link rot
Domain frontmatter
document audience classification
mojibake risk in canonical docs

Troubleshooting

AI cannot find Revit tools

Check:

npm run build has been run in MCP-Server.
The AI client's MCP config points to the correct MCP-Server/build/index.js.
The AI client has been restarted or has reloaded MCP servers.

MCP Server cannot connect to Revit

Check:

Revit is running.
The MCP service is enabled in the Revit ribbon.
localhost:8964 is not occupied.
If HTTP.sys / PID 4 is holding the port, try:

powershell

powershell -ExecutionPolicy Bypass -File scripts\release-port.ps1

Revit does not show the MCP Tools panel

Confirm the .addin file and DLL were deployed under the matching %APPDATA%\Autodesk\Revit\Addins\{version} directory, then restart Revit.

Important Rules

Keep one MCP/RevitMCP.csproj.
Keep one MCP/RevitMCP.addin.
Do not create version-specific .csproj or .addin files.
Do not create nested MCP/MCP/ directories.
Do not change .addin <Assembly> to an absolute path.
Do not convert Domain files to English-only.
Do not bypass the MCP server with hand-written WebSocket JSON.
For live Revit view, level, selection, or document state, AI must query live state in the current turn.

Document	Purpose
CLAUDE.md	Main AI agent constitution and project map
AGENTS.md	Redirect to `CLAUDE.md`
GEMINI.md	Redirect to `CLAUDE.md`
README.zh-TW.md	Traditional Chinese README
CONTRIBUTING.md	Contribution guide
CHANGELOG.md	Release history
domain/README.md	Domain SOP catalog
domain/lessons.md	Project lessons
.claude/skills/	AI skills
.claude/commands/	AI slash commands
scripts/README.md	Script documentation
docs/DOCUMENT_AUDIENCE_INVENTORY.md	Document audience inventory
docs/DOCS_STRUCTURE.md	Docs directory guide
log/README.md	Log append rules

License

MIT License

🛠️ Key Features

🚀 Use Cases

⚡ Developer Benefits

⚠️ Limitations

Revit MCP - AI-Powered Revit Control

What is this?

Quickstart (3 steps)

Current Project Counts

Architecture

Requirements

One-Click Setup

Install from MCP Registry

Manual Setup

1. Build the MCP Server

2. Build the Revit Add-in

3. Deploy the Add-in

AI Client Configuration

Switching Between AI Clients

Startup Flow

Project Structure

AI Docs and Human Docs

Domain, Skill, and Tool Responsibilities

QA/QC

Troubleshooting

AI cannot find Revit tools

MCP Server cannot connect to Revit

Revit does not show the MCP Tools panel

Important Rules

Document Navigation

License

Install