Zapcode-Figma-MCP
Zapcode is a tool that integrates with Figma to automate the design process. Users can easily create prototypes and receive feedback, enhancing design efficiency and promoting team collaboration. It features an intuitive interface that is user-friendly for beginners, but to fully utilize its capabilities, intermediate knowledge is required.
GitHub Stars
5
User Rating
Not Rated
Favorites
0
Views
28
Forks
0
Issues
0
ZapCode Figma MCP Server

Connect your Figma designs directly to your AI coding assistant! This tool acts as a bridge, allowing AI assistants that support the Model Context Protocol (MCP) to access real-time context from your selected Figma frames.
This server connects to the Zapcode Figma plugin to retrieve real-time context from your selected Figma frame, including its HTML/CSS structure, a generated image, and any exportable SVG assets. When invoked by an AI model, it provides this rich context, automatically extracts any SVG assets from the selection, saves them to your local workspace, and informs the LLM of their location.
Quick Install & Usage
This server is designed to be run directly via npx, requiring no permanent installation.
Prerequisites
- Node.js: Ensure you have Node.js (v16 or higher) installed. You can download it from nodejs.org.
- ZapCode Figma Plugin: You must install the companion plugin from the Figma Community.
- Installation Link: Zapcode - Figma Plugin
Integration with your AI Assistant
Configure your AI assistant to launch the server automatically. Here are some examples:
GitHub Copilot
In your Copilot configuration, add the following:
{
"servers": {
"Zapcode stdio mcp": {
"type": "stdio",
"command": "npx",
"args": ["zapcode-figma-mcp"]
}
},
"inputs": []
}
Claude for Desktop
Open your claude_desktop_config.json file and add the following to mcpServers:
- macOS/Linux:
~/Library/Application Support/Claude/claude_desktop_config.json - Windows:
%APPDATA%\Claude\claude_desktop_config.json
{
"mcpServers": {
"zapcode-figma-mcp": {
"command": "npx",
"args": ["zapcode-figma-mcp"]
}
}
}
```**Note:** Restart Claude for Desktop for the changes to take effect.
#### Other MCP Clients (e.g., Cline, Open WebUI)
For other clients, add a server configuration pointing to the `npx` command:
```json
{
"mcpServers": {
"zapcode-figma-mcp": {
"command": "npx",
"args": ["zapcode-figma-mcp"]
}
}
}
Usage Steps
- Run the ZapCode Plugin: Open your design file in the Figma desktop app and run the ZapCode plugin.
- Use Your AI Assistant: Select a frame or component in Figma. Go to your code editor and instruct your AI Assistant to perform a task. For example:
- "Generate a React component based on the current Figma selection."
- "Create HTML and CSS for the selected Figma frame."
- Automatic Context Fetching: Your AI assistant will automatically launch the
zapcode-figma-mcpserver, which will connect to the Figma plugin and fetch the necessary context to complete your request.
Additional Information
How It Works
Modern AI coding assistants often lack direct insight into your visual design process. This server solves that problem by:
- Running as a local process initiated by your MCP-compatible AI assistant.
- Establishing a connection with the ZapCode Figma Plugin when a request is made.
- Receiving context (structure, styles, assets) from the currently selected Figma frame.
- Serving this context to the AI assistant via the Model Context Protocol.
- Your AI assistant then uses this information to generate code that more accurately reflects your design.
Features
- Real-time Connection: Establishes a direct, local link between the ZapCode Figma Plugin and your AI Assistant.
- MCP Server: Hosts a local MCP server discoverable by compatible AI clients.
- Figma Context Tool: Provides the
get_figma_contexttool, enabling AI assistants to request data about the currently selected Figma frame(s). - Secure & Local: All communication happens on your machine; no design data is sent to external servers.
Tool Details: get_figma_context
This server exposes a single, powerful tool to the LLM.
- Description: Retrieves the selected Figma frame context. Saves any provided SVG assets to the local
assets/svgdirectory. - Input Schema:
{}(No input arguments are required). - Output: A combination of prompts and Figma context.
Troubleshooting
Error: Figma plugin is not connected.- This is the most common error. It means the MCP server is running but cannot find the Zapcode Figma plugin.
- Solution: Make sure you have opened the Figma desktop app and are running the Zapcode plugin on the design file you wish to query.
Port Conflict / EADDRINUSE Error- If you see an error that a port is already in use, it means another application is using port
3001or32896. - Solution: Stop the other application that is occupying the port.
- If you see an error that a port is already in use, it means another application is using port
Enterprise & Custom Integration
Boost your team's development speed and ensure brand consistency by integrating this system directly with your company's workflow.
We offer professional services to:
- Integrate the MCP server with your specific enterprise environment.
- Connect your proprietary design systems for seamless code generation.
- Develop custom features tailored to your team's needs.
Contact us to learn how we can help you accelerate your development lifecycle: subscribezapcode@gmail.com
License
This project is licensed under the MIT License.
0
Followers
0
Repositories
0
Gists
0
Total Contributions
Talk-to-Figma-Cursor is a JavaScript library that allows users to control Figma through voice commands. Utilizing voice recognition technology, it streamlines the design process. This tool is particularly beneficial for prototyping and collaboration, enabling hands-free operation. Users can input commands in natural language to easily manipulate elements within Figma.
figma-edit-mcp is a TypeScript library designed for efficiently editing Figma designs. Users can manipulate and customize design elements through an intuitive interface. It is particularly useful for automating design processes and improving workflows, offering a rich set of features for designers.