Overview
What is Webflow MCP?
Webflow MCP is a Node.js server implementing the Model Context Protocol (MCP) to enable AI agents to interact with Webflow APIs using the Webflow JavaScript SDK. It is intended for developers and AI tool users who want to manage Webflow sites, pages, and CMS content programmatically.
How to use Webflow MCP?
Install and configure by adding the npx command with your Webflow API token to your MCP client (e.g., Cursor or Claude Desktop). Prerequisites: Node.js, npm, and a Webflow account.
Key features of Webflow MCP
- List, get details, and publish sites
- List, get metadata, update pages, and page content
- List, create, and update CMS collections
- Create static, option, and reference fields
- Create, update, and publish collection items (staged or live)
- No prompts or resources currently implemented
Use cases of Webflow MCP
- Manage and publish Webflow sites via AI assistants
- Update page content and settings with natural language
- Create and modify CMS collections and items automatically
- Automate content updates across a Webflow project
- Retrieve site and page metadata for reporting
FAQ from Webflow MCP
What are the prerequisites to use Webflow MCP?
You need Node.js, npm, and a Webflow account with a valid API token.
How do I get a Webflow API token?
Go to Webflow's API Playground, log in, generate a token, and copy it from the Request Generator.
Does Webflow MCP support prompts or resources?
No, the current implementation only includes tools. Prompts and resources may be added in the future when broader MCP client support exists.
What if the server fails to start?
Ensure your Webflow token is valid and Node/npm are installed. Clear your npm cache or fix global package permissions if needed.
How do I run Webflow MCP in development mode?
Clone the repository, install dependencies, add your token to a .env file, and run npm start.
