figma-mcp-proxy
A read-only MCP proxy for the Figma REST API, providing tools to fetch design context, file data, images, comments, and user info using a personal access token. Designed for MCP clients that cannot use Figma's official OAuth flow.
README
figma-mcp-proxy
A small, read-only MCP compatibility proxy for Figma's REST API. It is intended for MCP clients that cannot complete the official Figma remote MCP OAuth flow.
This is not the official Figma MCP server. It returns raw Figma REST API data and does not provide Figma's generated framework code, design-system search, or write-to-canvas tools.
Requirements
- Node.js 18.18 or newer
- A Figma personal access token with the minimum read scopes needed for your files
Create a token in Figma account settings and keep it private. Each user supplies their own token; this project never contains or hosts tokens.
Run with npx
Until the package is published to the npm registry, run the public GitHub package directly:
npx --yes --package=github:dgehriger/figma-mcp-proxy -- figma-mcp-proxy
Set FIGMA_ACCESS_TOKEN in the environment used by your MCP client, then configure the server:
{
"mcpServers": {
"figma": {
"type": "local",
"command": "npx",
"args": [
"--yes",
"--package=github:dgehriger/figma-mcp-proxy",
"--",
"figma-mcp-proxy"
],
"env": {
"FIGMA_ACCESS_TOKEN": "${env:FIGMA_ACCESS_TOKEN}"
}
}
}
}
For GitHub Copilot Desktop, the file is %USERPROFILE%\\.copilot\\mcp-config.json. Set the token in the parent process environment before starting Copilot, or replace the environment reference with a locally managed secret mechanism supported by your installation. Do not commit the token to a config file.
Available tools
get_design_context— read a file or selected node from a Figma URL or file keyget_file— read a file documentget_file_nodes— read selected nodesget_images— request rendered image URLsget_comments— read file commentsget_current_user— verify the configured token
Agent guidance
The server sends a generalized Figma inspection workflow in the MCP initialization instructions, so an agent does not need a separate Figma skill to use this proxy correctly. The workflow teaches the agent to:
- request a direct design URL or file key and clarify the inspection goal;
- discover pages with shallow
get_filecalls before expanding a node tree; - inspect targeted pages, sections, and frames with
get_file_nodes; - identify concrete
FRAMEnodes before requesting context or images; - report node IDs, evidence, ordering, and access limitations; and
- avoid requesting credentials through chat.
Clients that support MCP prompts can invoke figma_inspection_workflow with a
goal and optional file URL or node ID. The same guidance is available as the
figma://workflow/inspection resource.
The guidance is intentionally generic. It does not assume a particular Figma team, project naming scheme, browser session, or organization.
Development
npm install
npm test
npm run check
The package is intentionally read-only. The official Figma remote MCP server remains the better option when a client can complete its OAuth flow.
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.