penpot

MCP server from Mart1M/penpot-mcp-server

Installation

Run this command in your terminal to add the MCP server to Claude Code.

Run in terminal:

Command

claude mcp add --transport stdio mart1m-penpot-mcp-server npx -y penpot-mcp-server@latest \
  --env PENPOT_API_URL="https://design.penpot.app/api" \
  --env PENPOT_ACCESS_TOKEN="your_penpot_token_here"

How to use

This MCP server integrates Penpot with AI assistants by exposing a set of tools that let you fetch and inspect Penpot boards and tokens from within an MCP client. The available tools are get_board, which retrieves a Penpot board and returns HTML, JSON, tokens used, and CSS variables; debug_board, which retrieves a board and returns the raw JSON for debugging; and get_tokens, which pulls design tokens from a Penpot file and returns DTCG-formatted tokens, CSS variables, and a summary. To use it, configure the Penpot access by supplying PENPOT_ACCESS_TOKEN and optionally PENPOT_API_URL (defaults to https://design.penpot.app/api) in your MCP client configuration. Then invoke the tools via the Penpot MCP server by pointing your MCP client to the penpot server name and providing the required parameters, such as the board URL for get_board and debug_board, or the file URL for get_tokens.

How to install

Prerequisites:

Node.js and npm installed on the machine where you will run the server.
Internet access to install dependencies from npm registry.

Install (development):

git clone <repository-url> or download the package.
Navigate to the project root.
Run:

npm install

Start the server:

npm start

Publish to npm (optional):

npm publish

Using via npx (production or integration):

npx -y penpot-mcp-server@latest

Notes:

Ensure you have the Penpot access token ready and set the PENPOT_ACCESS_TOKEN and PENPOT_API_URL environment variables in your MCP client's configuration.

Additional notes

Tips and common issues:

If you are running a self-hosted Penpot instance, provide PENPOT_API_URL to point to your API base URL; otherwise the default is https://design.penpot.app/api.
Ensure the access token has the necessary scope to read boards and tokens.
In MCP client configurations, you can opt to use npx for on-the-fly execution or run locally with node by providing the path to index.js as shown in the README.
When troubleshooting, enable verbose logging if available in your MCP client to capture tool responses and any API errors from Penpot.

Related MCP Servers

zen

1.1k

Selfhosted notes app. Single golang binary, notes stored as markdown within SQLite, full-text search, very low resource usage

MCP -Deepseek_R1

A Model Context Protocol (MCP) server implementation connecting Claude Desktop with DeepSeek's language models (R1/V3)

mcp-fhir

A Model Context Protocol implementation for FHIR

mcp

Inkdrop Model Context Protocol Server

mcp-appium-gestures

This is a Model Context Protocol (MCP) server providing resources and tools for Appium mobile gestures using Actions API..

dubco -npm

The (Unofficial) dubco-mcp-server enables AI assistants to manage Dub.co short links via the Model Context Protocol. It provides three MCP tools: create_link for generating new short URLs, update_link for modifying existing links, and delete_link for removing short links.