Server facts
f2c-mcp
Project
- Author
- f2c-ai
- Repository
- github.com/f2c-ai/f2c-mcp
- Website
- f2c.yy.com
- License
- MIT
- Category
- Design & Creative
Tech
- Runtime / language
- node
- Transports
- stdio
- Remote available
- No
- Authentication
- Not specified
Activity
- GitHub stars
- 411
- Last commit
- Data as of
Tags
- Figma
- ai coding assistant
- css
- d2c
- f2c
- html
- mcp
Overview
UI development from Figma can start with a node link or a selection in the desktop app. In f2c-mcp, get_code generates interface code for the chosen node; omitting a node ID uses the current selection. Node URLs can also supply the target ID. The documented output options include HTML/CSS and React, with CSS Modules and Tailwind CSS support. Figma design context is made available to AI coding tools, connecting the source design with requests made in an MCP-enabled IDE.
Image work follows the same node-targeting approach through get_image, which generates an image for the specified or selected node. Exported design images provide a reference for visually comparing generated interfaces with the Figma original. The project also describes downloading remote Figma images into local assets. A typical workflow is to configure the server in a supported IDE, open its chat, paste a Figma node link, and ask for code or images. The README notes that Figma REST API rate limits can affect operation and directs affected users to @f2c/mcp-plugin.
f2c-mcp setup notes
- f2c-mcp records stdio only; confirm a network transport upstream before planning remote access.