penpot

This MCP server provides full manipulation capabilities for Penpot, enabling an AI assistant to create, modify, and manage Penpot designs programmatically. It exposes a rich set of tools for projects, pages, shapes, components, teams, sharing, comments, exports, fonts, webhooks, and more, allowing you to drive Penpot workflows through natural language or scripted prompts. To get started, run the Penpot MCP server (via Docker or another supported method) and connect your MCP client to the server name (penpot). Once running, you can issue commands like listing projects, creating files, adding pages, creating shapes, updating properties, and exporting assets, all through the MCP interface.

You’ll interact with tools grouped by resource type (e.g., Project & File Management, Page Management, Shape Creation, Shape Manipulation, Component System, Team & Collaboration, File Sharing, Comments & Feedback, Export & Media, Font Management, Webhooks, Search, Profile & Stats). The server supports batch operations, session-based updates with revision tracking, and real-time API integration with Penpot, enabling responsive, stateful design automation. For example, you can create a new Penpot file, add a page, place shapes, apply styles, create or update components, and export assets, all in a single automated workflow.

Prerequisites:

• Docker installed and running on your machine or host
• Access token for Penpot (PENPOT_ACCESS_TOKEN) and Penpot API URL (PENPOT_API_URL)

Option A: Run via Docker (recommended for HTTP server)

1. Pull and run the latest Penpot MCP Server image:

docker pull ghcr.io/zcube/penpot-mcp-server:latest
docker run -i \
  -e PENPOT_API_URL="https://design.penpot.app" \
  -e PENPOT_ACCESS_TOKEN="your-access-token-here" \
  ghcr.io/zcube/penpot-mcp-server:latest

2. The server will start and listen for MCP client requests. Use the configured PENPOT_API_URL and token to authenticate with Penpot.

Option B: Build from Source (Node.js CLI is available via npm)

1. Ensure Node.js is installed (v14+ recommended).
2. Clone the repository and install dependencies:

git clone https://github.com/zcube/penpot-mcp-server.git
cd penpot-mcp-server
npm install

3. Run the CLI locally (requires Penpot credentials via env or .env):

export PENPOT_API_URL="https://design.penpot.app"
export PENPOT_ACCESS_TOKEN="your-access-token-here"
node path/to/server.js

4. Alternatively install the package globally and run the CLI:

npm install -g @zcubekr/penpot-mcp-server
penpot-mcp-server

Note: Docker is the recommended option for HTTP deployments, while npm/node-based installation is suitable for local testing and CLI usage.

Environment variables:

• PENPOT_API_URL: Penpot API base URL (e.g., https://design.penpot.app)
• PENPOT_ACCESS_TOKEN: Personal access token for Penpot API Tips:
• Ensure the Penpot API URL is reachable from your runtime environment and the access token has sufficient permissions.
• When using Docker, you can map ports if you want to expose a local endpoint for MCP clients.
• The MCP server supports batch operations and revision tracking; use these features to keep designs in sync with AI-driven workflows.
• If you encounter authentication errors, verify token scope and that the URL is correct for your Penpot instance.
• The available tools are grouped by category (Projects, Pages, Shapes, Components, Teams, etc.). Use the appropriate tool names when composing MCP commands.