revit-ai-native
Provides tools for interacting with Autodesk Revit 2024, enabling AI agents to inspect documents and selections, query elements, create and modify walls, doors, windows, rooms, manage views and sheets, and execute parameter changes within transactional plans.
Click on "Install Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@revit-ai-nativeWhat walls are selected in the current Revit document?"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
Autodesk Native Agent Runtime for Revit 2024
A production-grade bridge between AI agents (Antigravity, Claude, etc.) and Autodesk Revit 2024, built on the Model Context Protocol (MCP).
Architecture
Antigravity / AI Agent
↓ (MCP over stdio)
MCP Server (TypeScript, Node.js)
↓ (Named Pipe, JSON envelopes)
Revit 2024 Add-in (.NET Framework 4.8)
↓ (ExternalEvent)
Main-thread Dispatcher → PlanExecutor
↓ (TransactionGroup)
Revit API (Document, Wall, FamilyInstance, ...)
↓ (Verification + Assertions)
Structured Result → MCP ResponseQuick Start
1. Prerequisites
Revit 2024 installed (required for add-in build)
.NET SDK (for building C# projects)
Node.js 18+ (for MCP TypeScript server)
PowerShell (for build/install scripts)
2. Set REVIT_2024_PATH
If Revit is not in the default location (C:\Program Files\Autodesk\Revit 2024), set the environment variable:
$env:REVIT_2024_PATH = "D:\REVIT\2024\Revit 2024"3. Build
.\scripts\build-all.ps14. Install Add-in
.\scripts\install-revit2024.ps15. Open Revit 2024
Open or create a Revit project. The add-in starts the named-pipe server automatically.
6. Start MCP Server
.\scripts\start-mcp.ps17. Configure Antigravity
Add the MCP server to your Antigravity configuration:
{
"mcpServers": {
"revit": {
"command": "node",
"args": ["e:\\revit-A.I Native\\apps\\agent-mcp\\dist\\index.js"]
}
}
}8. Get Status
Call revit_get_status from your agent. You should see:
{
"connected": true,
"activeDocumentTitle": "YourProject.rvt",
"busy": false
}9. Preview a Plan
Send a plan to revit_preview_plan:
{
"schemaVersion": "1.0",
"requestId": "test-001",
"description": "Create a 5000mm wall",
"document": { "strategy": "active_document" },
"units": "mm",
"coordinateSystem": "project",
"executionMode": "preview",
"operations": [...],
"safety": { "requireUserConfirmation": true }
}10. Approve & Commit
Call revit_commit_plan with the same plan. If confirmation is required, approve in Revit.
11. Inspect Audit Log
Audit logs are at %LOCALAPPDATA%\AutodeskNativeAgent\logs as JSONL files.
Supported Operations
Operation | Status |
| ✅ Supported |
| ✅ Supported |
| ✅ Supported |
| ✅ Supported |
| ✅ Supported |
| ✅ Supported |
| ✅ Supported |
| ✅ Supported |
| ✅ Supported |
| ✅ Supported |
| ✅ Supported |
| ✅ Supported |
| ✅ Supported |
| ✅ Supported |
| ✅ Supported |
| ✅ Supported |
| ✅ Supported |
| ✅ Supported |
MCP Tools
Tool | Description |
| Connection status and document info |
| Protocol version, methods, operations |
| Active document identity |
| Current selection |
| Structured element query |
| Validate without executing |
| Dry-run resolution |
| Execute in a transaction |
| Job status and result |
| Cancel queued/running job |
| Rollback completed job |
| Sanitized audit entries |
Key Design Decisions
No Revit API from background threads — All Revit API calls go through the ExternalEvent dispatcher
Atomic execution — Plans run inside a single TransactionGroup; failures roll back
Structured errors — Every failure carries a stable error code
Unit safety — All lengths carry explicit units; no hardcoded
/ 304.8Type safety — Type/level resolution is deterministic; no random
FirstElement()Path security — Export/SaveAs paths are canonicalized and checked for traversal
Confirmation tokens — Destructive operations require human approval
Documentation
License
MIT
This server cannot be installed
Maintenance
Resources
Unclaimed servers have limited discoverability.
Looking for Admin?
If you are the server author, to access and configure the admin panel.
Related MCP Connectors
OCR, transcription, file extraction, and image generation for AI agents via MCP.
Shared, governed long-term memory for AI agents across tools and sessions via MCP and REST.
MCP server for secureFlows: token-free URL builders and integration-linting tools for AI agents.
Latest Blog Posts
- Who's Calling? MCP Hosts Are an Identity Blind Spot (And the Spec Knows It)By Om-Shree-0709 on .mcpAgent IdentityOAuth 2.1
- Your AI Chatbot Just Exposed Your CEO's Salary to an InternBy Om-Shree-0709 on .Agent IdentityMCP SecurityOAuth Delegation
- Why MCP Servers Need Execution Sandboxing (And Why Your Current Stack Isn't Enough)By Om-Shree-0709 on .Agentic AiPrompt InjectionWebAssembly
MCP directory API
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/tuanank2112-crypto/revit-ai-native'
If you have feedback or need assistance with the MCP directory API, please join our Discord server