Loading...
List files, inspect components, read comments, and export variables via Figma REST API
175+ deploysYour AI agent gets 5 capabilities in one connection. Deploy once, use from any client.
get_file
Fetch Figma file tree. Optional query params: 'ids' (comma-separated node IDs to subset), 'depth' (1-4 for partial tree), 'geometry=paths' (include vector paths). Returns document.children[] with pages, frames, components, each having id, name, type, absoluteBoundingBox, fills, strokes.
https://api.figma.com/v1/files/{file_key}
list_components
List published components in a file. Returns meta.components[] with key, name, description, containing_frame (name, nodeId, pageName), and created/updated timestamps. Use component key to fetch instances across files.
https://api.figma.com/v1/files/{file_key}/components
get_comments
Get all comments on a file. Optional: 'as_md=true' for Markdown formatting. Returns comments[] with id, message, created_at, resolved_at, user (handle, img_url), client_meta (node_id, node_offset for pin coordinates), and order_id for threading replies.
https://api.figma.com/v1/files/{file_key}/comments
get_variables
Fetch local variables (design tokens). Returns meta.variables{} keyed by ID, each with name, resolvedType (COLOR, FLOAT, STRING, BOOLEAN), valuesByMode{} mapping mode IDs to values, variableCollectionId, and scopes[] (ALL_SCOPES, CORNER_RADIUS, etc). Requires 'variables:read' scope.
https://api.figma.com/v1/files/{file_key}/variables/local
list_projects
List all projects in a team. Returns projects[] with id and name. Use project ID with GET /v1/projects/{id}/files to list files within. Requires team_id from your Figma team URL (figma.com/files/team/{team_id}).
https://api.figma.com/v1/teams/{team_id}/projects
Connect Figma to Claude Code with one command:
claude mcp add figma https://datafaucet.dev/api/mcp/t/figma{notion_token}) with your real credentialsNeed API keys? Try our free sandbox templates first to see MCP in action with zero setup.
Replace these placeholders with your real values before deploying:
{figma_token}API settings page{file_key}Your value{team_id}Resource identifierPreview:
GET https://api.figma.com/v1/files/{file_key} X-Figma-Token: {figma_token}
claude mcp add figma --url https://datafaucet.dev/api/mcp/t/figma
Paste into your client's MCP configuration file · most users connect in under 60s
claude mcp add figma --url https://datafaucet.dev/api/mcp/t/figmaWhat your AI agent sees
I have access to 5 tools from Figma: get_file, list_components, get_comments, and 2 more.
“Fetch Figma file tree. Optional query params: 'ids' (comma-separated node IDs to subset), 'depth' (1-4 for partial tree), 'geometry=paths' (include vector paths). Returns document.children[] with pages, frames, components, each having id, name, type, absoluteBoundingBox, fills, strokes.”
Calling get_file... Done. Here are the results.
Try these prompts
“List all files in our team project”
Copy“Show me the components in the design system file”
Copy“Get all comments on the homepage redesign file”
CopyClick any prompt to copy, then paste into your AI client.