
Webflow
com.webflowPublished by webflow.comOfficial publisherv2.0.0Updated Oct 24, 2026
AI-powered design and management for Webflow Sites
Overview
Webflow's remote MCP server lets an AI assistant read and edit Webflow sites, blog content and Designer pages through OAuth-authorized Webflow APIs.
- What you can do
- Analyze and manage Webflow site content, including blog posts, topic ideas and internal links between posts via the Webflow Data API and JavaScript SDK (R3, R29). Create and edit page elements such as hero sections, cards and CTA buttons with responsive design through the Designer bridge app (R29, R72). Open a site in the Webflow Designer and get a link to it from the assistant (R23, R38). Retrieve the local connection URL for the Designer bridge app (R76).
- When to use it
- Useful when you manage Webflow sites and want an assistant to draft or reorganize blog content, suggest SEO topics, or build and adjust page sections inside the Webflow Designer rather than editing by hand.
- Requirements
- Remote endpoint at mcp.webflow.com with OAuth login to authorize which Webflow sites the server may access (R9, R21). A Webflow account (R6). Node.js 22.3.0 or higher for the client-side setup, including the npx mcp-remote bridge in Claude Desktop (R11, R34). The published Webflow MCP Bridge App opened in the Designer for canvas work (R26). The local install option instead needs a self-registered bridge app with Admin permissions and a Webflow API token in WEBFLOW_TOKEN (R51, R69).
Installation
In SourceWeft
- Open Webflow in the dashboard and add it to a workspace.
- Connect the required credentials. They are stored privately and never shown on public pages.
- Enable the server for the chats that should use its tools.
Web executable via Streamable HTTP. Remote servers run from the web runtime once configured in a workspace.
Use with an agent
Paste this prompt into Claude Code, Codex, Cursor or another agent — or into SourceWeft chat. The agent reads this server's install guide, shows you its publisher, endpoint and sign-in, and adds it to your client once you agree. Keys never go into a command.
Read https://sourceweft.com/mcp/com.webflow%2Fmcp/install.md and connect the MCP server it describes to my client. Before changing any config, show me its publisher, endpoint and sign-in method and wait for my OK. Never put an API key in a command or in a file I would commit; ask me where it should come from.Other MCP clients
Add this to your client's mcpServers config. The client completes OAuth sign-in on first connect.
{
"mcpServers": {
"webflow": {
"type": "http",
"url": "https://mcp.webflow.com/mcp"
}
}
}README
Webflow's MCP server
A Node.js server implementing Model Context Protocol (MCP) for Webflow using the Webflow JavaScript SDK. Enable AI agents to interact with Webflow APIs. Learn more about Webflow's Data API in the developer documentation.
Prerequisites
🚀 Remote installation
Get started by installing Webflow's remote MCP server. The remote server uses OAuth to authenticate with your Webflow sites, and a companion app that syncs your live canvas with your AI agent.
Requirements
- Node.js 22.3.0 or higher
Note: The MCP server currently supports Node.js 22.3.0 or higher. If you run into version issues, see the Node.js compatibility guidance.
Cursor
Add MCP server to Cursor
- Go to
Settings → Cursor Settings → MCP & Integrations. - Under MCP Tools, click
+ New MCP Server. - Paste the following configuration into
.cursor/mcp.json(or add thewebflowpart to your existing configuration):
Tip: You can create a project-level
mcp.jsonto avoid repeated auth prompts across multiple Cursor windows. See Cursor’s docs on configuration locations.
- Save and close the file. Cursor will automatically open an OAuth login page where you can authorize Webflow sites to use with the MCP server.
Open the Webflow Designer
- Open your site in the Webflow Designer, or ask your AI agent:
Open the MCP Webflow App
- In the Designer, open the Apps panel (press
E). - Launch your published "Webflow MCP Bridge App".
- Wait for the app to connect to the MCP server.
Write your first prompt
Try these in your AI chat:
Claude desktop
Add MCP server to Claude desktop
- Enable developer mode:
Help → Troubleshooting → Enable Developer Mode. - Open developer settings:
File → Settings → Developer. - Click
Get Startedor edit the configuration to openclaude_desktop_config.jsonand add:
- Save and restart Claude Desktop (
Cmd/Ctrl + R). An OAuth login page will open to authorize sites.
Open the Webflow Designer
- Open your site in the Webflow Designer, or ask your AI agent:
Open the MCP Webflow App
- In the Designer, open the Apps panel (press
E). - Launch your published "Webflow MCP Bridge App".
- Wait for the app to connect to the MCP server.
Write your first prompt
Reset your OAuth token
To reset your OAuth token, run the following command in your terminal.
Node.js compatibility
Please see the Node.js compatibility guidance on Webflow's developer docs.
Local Installation
You can also configure the MCP server to run locally. This requires:
- Creating and registering your own MCP Bridge App in a Webflow workspace with Admin permissions
- Configuring your AI client to start the local MCP server with a Webflow API token
1. Create and publish the MCP bridge app
Before connecting the local MCP server to your AI client, you must create and publish the Webflow MCP Bridge App in your workspace.
Steps
-
Register a Webflow App
- Go to your Webflow Workspace and register a new app.
- Follow the official guide: Register an App.
-
Get the MCP Bridge App code
- Option A: Download the latest
bundle.zipfrom the releases page. - Option B: Clone the repository and build it:
- Then build the project following the repository instructions.
- Option A: Download the latest
-
Publish the Designer Extension
- Go to Webflow Dashboard → Workspace settings → Apps & Integrations → Develop → Your App.
- Click “Publish Extension Version”.
- Upload your built
bundle.zipfile.
-
Open the App in Designer
- Once published, open the MCP Bridge App from the Designer → Apps panel in a site within your workspace.
2. Configure your AI client
Cursor
Add to .cursor/mcp.json:
Claude desktop
Add to claude_desktop_config.json:
3. Use the MCP server with the Webflow Designer
- Open your site in the Webflow Designer.
- Open the Apps panel (press
E) and launch your published “Webflow MCP Bridge App”. - Wait for the app to connect to the MCP server, then use tools from your AI client.
- If the Bridge App prompts for a local connection URL, call the
get_designer_app_connection_infotool from your AI client and paste the returnedhttp://localhost:<port>URL.
Optional: Run locally via shell
Reset your OAuth Token
To reset your OAuth token, run the following command in your terminal.
Node.js compatibility
Please see the Node.js compatibility guidance on Webflow's developer docs.
❓ Troubleshooting
If you are having issues starting the server in your MCP client e.g. Cursor or Claude Desktop, please try the following.
Make sure you have a valid Webflow API token
- Go to Webflow's API Playground, log in and generate a token, then copy the token from the Request Generator
- Replace
YOUR_WEBFLOW_TOKENin your MCP client configuration with the token you copied - Save and restart your MCP client
Make sure you have the Node and NPM installed
Run the following commands to confirm you have Node and NPM installed:
Clear your NPM cache
Sometimes clearing your NPM cache can resolve issues with npx.
Fix NPM global package permissions
If npm -v doesn't work for you but sudo npm -v does, you may need to fix NPM global package permissions. See the official NPM docs for more information.
Note: if you are making changes to your shell configuration, you may need to restart your shell for changes to take effect.
🛠️ Available tools
See the ./tools directory for a list of available tools
🗣️ Prompts & resources
This implementation doesn't include prompts or resources from the MCP specification. However, this may change in the future when there is broader support across popular MCP clients.
📄 Webflow developer resources
⚠️ Known limitations
Static page content updates
The pages_update_static_content endpoint currently only supports updates to localized static pages in secondary locales. Updates to static content in the default locale aren't supported and will result in errors.
Source: README.md at commit c4fa339
Tools
0Version history
1- v2.0.0LatestSep 16, 2026
Score
85% of the applicable checks met
How complete this listing is, computed from the registry entry, the endpoint and public sources. Not a rating of the server and not a statement that it is safe or working today: reachability is a dated check, and no installs or reviews go into the score.
- Reached in the last 30 daysRequired25 pts
- Auth declaredRequired10 pts
- Tools listed15 pts
Answer tools/list without a credential, or ship a package that starts in a sandbox without one.
- DescriptionRequired15 pts
- Logo10 pts
- Documentation15 pts
- License10 pts
