About Figma MCP
Figma MCP gives design and development workflows access to Figma REST API data. It lists a file's top-level frames and saves a selected frame's PNG, node JSON, and image-fill map locally.
Key features
- 01List and filter top-level Figma frames
- 02Inspect frame names, IDs, dimensions, and pages
- 03Fetch a frame's rendered PNG
- 04Save node JSON and normalized image-fill data
- 05Keep fetched files beneath a configured output root
Use cases
- 01Review the frames available in a Figma file
- 02Retrieve a frame's image and structured node data
- 03Save frame assets locally for downstream inspection
MCP support
Exposes two tools to MCP clients: one lists top-level frames, and the other fetches a frame and writes its PNG and JSON data to a local output directory. It uses a Figma personal access token.
Configuration
Local server config is the same for every client. Copy it into your client’s MCP config file.
We haven’t collected a package or remote endpoint for this server. Follow the instructions in the project README.
Tools
2Saved tool list · updated 2w ago
| Parameter | Type | Required | Description |
|---|---|---|---|
| query | string | No | Optional case-insensitive page or frame-name substring. |
| file_key | string | Yes | File key parsed from the Figma URL. |
Details
From the project READMEFigma MCP
Figma REST API MCP server. 2 tools list the top-level frames in a file and fetch one frame's rendered PNG plus its node JSON and image-fill map into a local output directory.
Install
Claude Code:
claude mcp add figma -e FIGMA_TOKEN=your-token -- npx -y @scalably-io/figma-mcp
Codex:
codex mcp add figma --env FIGMA_TOKEN=your-token -- npx -y @scalably-io/figma-mcp
Claude Desktop: download figma-mcp.mcpb from the latest GitHub release and open it.
Setup
- Create a personal access token at figma.com under account settings, with
file_content:readscope. - Find the file key in any Figma file URL: it is the segment after
/design/or/file/, for examplehttps://www.figma.com/design/AbC123xyz/My-filegivesAbC123xyz. Both tools take it asfile_key. - If
FIGMA_TOKENis not set the server still starts and lists its tools; every call then fails withFIGMA_NOT_CONFIGUREDuntil the token is provided. fetch_frameis the only tool that writes files (so it is not marked read-only); it writesframe.png,node.json, andfills.jsoninto a directory below the configured output root (default./figma-output). The other tool is fully read-only.
Tools (2)
| Tool | What it does |
|---|---|
list_frames | List and optionally filter all top-level FRAME nodes in a Figma file, with exact page/name/node-id data, dimensions, and file metadata |
fetch_frame | Fetch one Figma frame and transactionally write frame.png, node.json, and a normalized image-fill map into a directory below the output root |
Configuration
| Variable | Required | Purpose |
|---|---|---|
FIGMA_TOKEN | yes | Figma personal access token with file_content:read |
FIGMA_API_BASE_URL | no | Override the Figma REST API base URL (default https://api.figma.com/v1) |
FIGMA_REQUEST_TIMEOUT_MS | no | Per-request timeout in milliseconds (default 60000) |
FIGMA_MAX_JSON_BYTES | no | Upper bound on a Figma JSON response, in bytes (default 104857600) |
FIGMA_MAX_FRAME_BYTES | no | Upper bound on a downloaded frame PNG, in bytes (default 104857600) |
FIGMA_OUTPUT_ROOT | no | Directory that every fetch_frame output directory must stay below (default ./figma-output, created on first use) |
FIGMA_ALLOW_HTTP_FOR_TESTS | no | Test suite only: 1 allows a plain-HTTP loopback API base. Never set it in normal use |
Reply shape
Every tool returns plain JSON with status (succeeded, partial, no_op), summary, target, result, proof, warnings, recovery. Failures throw a plain error string: <code>: <message> <hint>.
Limits
fetch_frame renders through Figma's image API, so very large frames or extreme scale values can exceed the configured JSON or frame byte limits; lower scale or split the frame if that happens. fills.json URLs are temporary and expire within 14 days.
Verify
Each release lists the package version, the .mcpb sha256 and the production commit it was derived from in CHANGELOG.md. CI runs the tests and a clean install of the packed tarball on every push.
Privacy Policy
This server runs locally, on your machine, under your own credentials. It collects no personal data, contains no telemetry, stores nothing persistently beyond the frame bundles you explicitly fetch, and talks only to the vendor API it wraps. No third party, including Scalably, receives your data. Contact: [email protected]. Canonical copy: https://scalably.io/connector-privacy.html
License
MIT. Copyright Scalably.
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.