Webflow's MCP server now supports the Designer
Webflow’s MCP server now supports the Designer API, enabling AI agents to interact directly with the Webflow Designer canvas in real-time. This pivotal update expands the server’s capabilities beyond content management, opening new possibilities for AI-assisted visual design.
What’s new
The MCP server now includes tools that connect directly to the Webflow Designer, allowing AI agents to:
- Create and modify design elements on the canvas
- Manage styles, variables, and components
- Work with responsive breakpoints and layouts
- Access real-time design data and structure
Companion app for Designer connectivity

A new companion app automatically installs during authorization and enables seamless communication between your AI agent and the Designer:
- Automatic setup: The companion app installs automatically when you authorize the MCP server
- Real-time sync: Maintains a persistent connection between your AI agent and active design sessions
- Designer access: Must remain open in the Designer for MCP tools to function
Companion app connectivity
Keep the companion app open: The MCP Companion App must remain open in the Webflow Designer for Designer API tools to function. Close the app, and you’ll lose access to canvas-based operations.
Getting started
To use the new Designer API features:
- Update your MCP configuration following the installation guide
- Authorize your sites through the OAuth flow (the companion app installs automatically)
- Open the Designer and launch the “Webflow MCP Companion App” from the Apps panel
- Start designing with AI-powered prompts like “Create a hero section with responsive design”
Important requirements
Node.js version 22.3.0+
The MCP server now requires Node.js version 22.3.0 or higher. If you’re using an older version:
- Use
nvm
to install and switch to Node.js 22.3.0 - Clear your
npx
cache:rm -rf ~/.npm/_npx
- Restart your AI client after updating
See the Node.js compatibility section for detailed troubleshooting steps.
Learn more
For complete setup instructions, tool documentation, and troubleshooting guides, see our MCP server and AI tools documentation.