Figma MCP server
Inspect selected Figma nodes and published library assets with Arc0’s tools. Agents can render nodes and post design feedback, while file reads stay deliberately small. Connect through Arc0 to use the actions your identity is allowed to call.
Working with Figma
A design review is more useful when it starts with a specific frame. Give the agent the file key and a few node IDs for figma.get_file_nodes. The tool reads one to five nodes at limited depth. figma.get_file can inspect a small file, but large response bodies may fail; narrowing the selection avoids pretending that an entire complex file was examined.
figma.export_images renders selected nodes for a visual comparison. A null render is not an image, and generated URLs expire. For a component audit, figma.list_file_components and figma.list_file_styles inventory published assets in a main file. They do not include unpublished work or accept branch keys for that inventory.
Read existing feedback with figma.list_comments before adding a new observation through figma.post_comment. Comments are visible to collaborators and may notify them, so review the text before posting. figma.delete_comment can remove only a comment authored by the connected user. After a lost posting response, inspect the comment list before adding the same feedback again.
Official MCP reference: Figma documentation.
Agent workflows for Figma.
Inspect a specific frame
Use figma.get_file_nodes for a few known node IDs rather than requesting an entire design file.
Render a review image
Use figma.export_images to obtain a PNG, JPG, SVG, or PDF render of selected nodes.
Read and leave feedback
Use figma.list_comments before figma.post_comment adds feedback visible to file collaborators.
Inventory published assets
Use figma.list_file_components and figma.list_file_styles for a known main-file library.
8 Figma actions, graded by risk.
These actions come from the Arc0 registry. Your identity, provider permissions and Arc0 policies determine which actions can run.
read
6Look things up, subject to your permissions.
- figma.get_fileGet file
Use when inspecting a small file at depth 1–2; not for bulk reads. Prefer get_file_nodes with few IDs. Tier 1 View/Collab budget: up to 20/month; honor long Retry-After. Whole-file asset maps remain; raw bodies over 1 MiB fail (502 provider_response_too_large); Figma may return 400.
- figma.get_file_nodesGet file nodes
Use when inspecting 1–5 node IDs at depth 1–2; not for bulk file export. Tier 1 View/Collab budget: up to 20/month; honor long Retry-After. Null tree means missing. Raw bodies over 1 MiB fail (502 provider_response_too_large); use fewer IDs or less depth. Figma may return 400.
- figma.export_imagesExport images
Use when rendering up to 20 nodes as PNG, JPG, SVG or PDF. URLs expire in 30 days; null means no render. Conserve Tier 1 requests. Very large renders may return 500: use fewer IDs or a lower scale. Do not use for bulk full-file export.
- figma.list_commentsList comments
Use when reading feedback on one known file or checking an uncertain comment write. Do not use for broad activity polling or author-profile lookup. Author profiles are omitted.
- figma.list_file_componentsList file components
Use when inventorying published components in a known file library. Supply a main-file key, not a branch key. Unpublished assets are not included.
- figma.list_file_stylesList file styles
Use when inventorying published styles in a known file library. Supply a main-file key, not a branch key. Unpublished assets are not included.
write
1Create and change things. Allow, or ask the user first.
- figma.post_commentPost comment
Use when sharing feedback or replying to a root comment. Every collaborator on the file sees it and may be notified. Do not use for private notes, bulk or speculative posting. No idempotency: after an unknown outcome, list comments and read back before retrying.
destructive
1Delete, cancel or archive. Ask first, or deny outright.
- figma.delete_commentDelete comment
Use when removing a specific comment authored by the connected user, including test cleanup. Confirm its id before deletion; other authors’ comments cannot be deleted.
Control access to Figma.
Connect the intended Figma account, authorize your AI client and set policies for its tools.
Provider permissions and Arc0 policy both apply. Allow reads, require approval for selected writes, or deny destructive actions. The audit log records execution decisions.
Configure policies and approvalsExample policy choices
- Read actions: allow where authorized
- figma.post_comment: require approval
- figma.delete_comment: deny
Arc0 MCP endpoint: https://api.arc0.ai/mcp
How Figma connects.
Connect Figma using API key. OAuth grants the provider scopes shown during consent; API keys and other credentials retain the permissions configured at the provider. Arc0 stores credentials in its encrypted vault and does not pass them to the model. Disconnect to remove access through Arc0; revoke a key at the provider to invalidate it there. Token refresh and provider-side revocation depend on the authentication lane and provider.
The connection serves authorized calls through Arc0 MCP and REST. How Arc0 handles credentials →
- AUTH
- API key
- CREDENTIALS
- Per-tenant encrypted vault
- MODEL SEES
- Results only, never credentials
- AUDIT LOG
- Every call, on every plan
Use Figma with your AI client.
Sign in to Arc0 and connect Figma in My apps. Then authorize your AI client separately. It only receives tools and accounts allowed for that identity.
Figma with Claude
Claude Desktop + Claude Code
- In Claude Desktop, add Arc0 through the remote MCP settings and use OAuth. Follow the generic remote-client steps if your account’s interface differs.
- For Claude Code, run the command below. Open /mcp, select Arc0, and authenticate in your browser. Select your workspace and review access.
- Check the connection with claude mcp get arc0.
claude mcp add --transport http --scope user arc0 https://api.arc0.ai/mcpFull MCP setup and troubleshootingFigma with ChatGPT
ChatGPT on the web
- Open Plugins in ChatGPT, select the plus button, then Add custom MCP server.
- Name it Arc0, enter the server URL below, and choose OAuth. Use dynamic client registration if asked; no Arc0 API key is needed.
- Review the notice and select Create as a plugin. Complete Arc0 sign-in, workspace selection and consent, then install and select the plugin in your conversation. Labels and access vary by account; managed workspaces may require administrator approval.
https://api.arc0.ai/mcpFull MCP setup and troubleshootingFigma with OpenAI Codex
Codex app + CLI
- Run the registration command below. The app and CLI share ~/.codex/config.toml.
- Complete browser OAuth, choose your Arc0 workspace, and review access. If authentication does not start, run codex mcp login arc0.
- Check codex mcp list, then open a new Codex chat. Inspect an existing arc0 entry before changing it.
codex mcp add arc0 --url https://api.arc0.ai/mcpFull MCP setup and troubleshootingFigma with Cursor
Remote MCP setup
- Add Arc0 in your client’s remote MCP settings using the server URL below.
- Choose OAuth, sign in to Arc0, select the intended workspace and review access.
- Confirm the server is connected and start with an allowed read. These are generic remote-MCP steps; consult your client’s settings for its current interface.
https://api.arc0.ai/mcpFull MCP setup and troubleshootingFigma with VS Code
Remote MCP setup
- Add a remote HTTP MCP server named Arc0 using the URL below.
- Use OAuth and finish Arc0 sign-in, workspace selection and consent.
- Confirm discovery succeeds before requesting a read. These are generic remote-MCP steps; client UI and availability may differ.
https://api.arc0.ai/mcpFull MCP setup and troubleshootingFigma with Any MCP client
Streamable HTTP
- Use a client that supports remote MCP over Streamable HTTP. Add the server URL below and choose OAuth.
- For a client without OAuth, use a self-bound API key with execute permission in the X-API-Key header. Store it in secret storage or an environment variable, outside source control. Never send both an API key and an OAuth bearer token.
- Use the MCP clients guide for the SDK transport example and Embedded identity for business-managed end users.
https://api.arc0.ai/mcpFull MCP setup and troubleshootingAfter setup, ask your agent to find an allowed read action, load its schema, and execute it on the intended account. Check Activity for the receipt. A connected MCP client alone does not confirm a successful provider call.
More design and images apps.
Figma and Arc0, answered.
Is there an official Figma MCP server, and how does Arc0 differ?
Yes. Figma documents remote and desktop MCP servers. Arc0’s Figma tools use REST for bounded node inspection, rendering, published assets, and comments; they do not expose canvas editing.
Which permissions does Arc0 need for Figma?
Use a personal access token with current_user:read for identity, file_content:read, file_comments:read, file_comments:write and library_content:read for the listed tools. The current version does not use OAuth or plan access tokens.
Does this work with Claude, ChatGPT, Codex and Cursor?
Yes, through Arc0 MCP when your Claude, ChatGPT, Codex, or Cursor client supports remote MCP and your account permits it. Connect the app in Arc0 and configure your client using the Arc0 setup guide. App permissions still apply.
How are credentials stored, refreshed and revoked?
Arc0 stores the PAT encrypted and cannot refresh it. Replace it when it expires. Disconnect removes Arc0’s local access; revoke the PAT in Figma’s Security settings to invalidate it at Figma.
What happens with rate limits or errors?
Figma limits depend on endpoint tier, plan, and seat. Wait times can be long. Arc0 reports throttling and oversized responses; reduce node count or depth for large reads. Check comments before repeating a post with an unknown outcome.
What is the availability of these Figma tools?
These tools are available. Connect the app in Arc0 and authorize your AI client. The action list describes the current implementation.
Plug Figma into your agent.
Figma MCP server: 8 actions including get file, get file nodes, export images. Connect through Arc0.
Free to build · MCP + REST · Audit log on every plan
Figma and its logo are trademarks of their respective owners. Arc0 is not affiliated with or endorsed by Figma.