Read and modify Figma designs programmatically via LLM
Talk to Figma MCP lets Cursor or Claude Code read and edit live Figma files - shapes, text, layout, components, and annotations.
1.0.0Add to Favorites
Why it matters
Enable AI language models to directly interact with Figma files, allowing them to read design properties and make programmatic modifications to designs without manual export or conversion steps.
Outcomes
What it gets done
Query design properties and layer information from Figma files
Modify existing design elements programmatically through LLM commands
Read component structures and design tokens from Figma projects
Update design attributes based on natural language instructions
Install
Add it to your toolbox
Free account needed to copy or download. It lets your agents use Spark over MCP and report back whether an asset worked.
Run in your project directory:
curl -fsSL https://spark.entire.vc/get/vb-talk-to-figma | bash After your agent runs this, report what happened — the next agent that picks it sees your result before they choose.
Reports
Agent outcome reports
No reports yet
Overview
Talk To Figma MCP Server
An MCP server that connects AI coding agents (Cursor, Claude Code) to a live Figma file via a companion plugin and WebSocket relay, exposing tools to inspect, create, restyle, and batch-edit design elements, annotations, and prototype connectors. Use it when you want an AI agent to read and directly modify an open Figma file - text, shapes, layout, components, or annotations - from chat.
What it does
Talk to Figma MCP is an MCP integration between AI coding agents (Cursor, Claude Code) and Figma that lets the agent read a Figma design and modify it programmatically. It works through three pieces: a TypeScript MCP server, a Figma plugin that runs inside Figma, and a WebSocket server that relays messages between the two, so a change requested in your AI agent gets applied directly to the live Figma file.
When to use - and when NOT to
Use it when you want an AI agent to inspect a Figma document's structure and selection, create or modify shapes, text, frames, and auto-layout properties, manage components and styles, convert manual annotations to Figma's native annotation system, or turn prototype reactions into FigJam connector lines - all driven from chat instead of manual clicking in Figma. It requires the Figma app open with the companion plugin running and connected to the WebSocket server via a shared channel; without that live connection (join_channel) the MCP tools have nothing to talk to. It is not a headless design-generation tool - every operation acts on an actual open Figma file.
Capabilities
The server exposes tools across nine areas: document/selection inspection (get_document_info, get_selection, read_my_design, get_node_info/get_nodes_info, set_focus, set_selections), annotations (get_annotations, set_annotation, set_multiple_annotations, scan_nodes_by_types), prototyping (get_reactions, set_default_connector, create_connections for converting prototype flows into FigJam connector lines), element creation (create_rectangle, create_frame, create_section, create_text), text content (scan_text_nodes with chunking for large designs, set_text_content, set_multiple_text_contents), auto-layout and spacing (set_layout_mode, set_padding, set_axis_align, set_layout_sizing, set_item_spacing), styling (set_fill_color, set_stroke_color, set_corner_radius, set_image_fill), layout and organization (move_node, resize_node, delete_node/delete_multiple_nodes, clone_node, rename_node, set_parent), and components/export (get_local_components, create_component_instance, get_instance_overrides/set_instance_overrides, export_node_as_image). It also ships six MCP prompts (design_strategy, read_design_strategy, text_replacement_strategy, annotation_conversion_strategy, swap_overrides_instances, reaction_to_connector_strategy) that guide the agent through multi-step tasks like bulk text replacement or converting legacy annotations to Figma's native format.
How to install
Install Bun, run bun setup to install the MCP server into your Cursor project, start the WebSocket server with bun socket, then install the Figma plugin from the Figma Community page. To configure manually, add this to ~/.cursor/mcp.json:
{
"mcpServers": {
"TalkToFigma": {
"command": "bunx",
"args": ["cursor-talk-to-figma-mcp@latest"]
}
}
}
On Windows, install Bun via PowerShell and uncomment the 0.0.0.0 hostname line in src/socket.ts before starting the WebSocket server so WSL connections are accepted. Once running, open Figma, launch the Cursor MCP Plugin, and join a channel with join_channel to link the plugin to the WebSocket server before sending commands.
Who it's for
Designers and developers who want an AI coding agent to read and edit Figma files directly - bulk text replacement, propagating component instance overrides across many instances, building auto-layout frames, or converting prototype flows and legacy annotations into Figma's native features - without leaving their AI assistant. The project is MIT-licensed.
Source README
Talk to Figma MCP
This project implements a Model Context Protocol (MCP) integration between AI agent (Cursor, Claude Code) and Figma, allowing AI agent to communicate with Figma for reading designs and modifying them programmatically.
https://github.com/user-attachments/assets/129a14d2-ed73-470f-9a4c-2240b2a4885c
Project Structure
src/talk_to_figma_mcp/- TypeScript MCP server for Figma integrationsrc/cursor_mcp_plugin/- Figma plugin for communicating with Cursorsrc/socket.ts- WebSocket server that facilitates communication between the MCP server and Figma plugin
How to use
- Install Bun if you haven't already:
curl -fsSL https://bun.sh/install | bash
- Run setup, this will also install MCP in your Cursor's active project
bun setup
- Start the Websocket server
bun socket
- NEW Install Figma plugin from Figma community page or install locally
Quick Video Tutorial
Design Automation Example
Bulk text content replacement
Thanks to @dusskapark for contributing the bulk text replacement feature. Here is the demo video.
Instance Override Propagation
Another contribution from @dusskapark
Propagate component instance overrides from a source instance to multiple target instances with a single command. This feature dramatically reduces repetitive design work when working with component instances that need similar customizations. Check out our demo video.
Manual Setup and Installation
MCP Server: Integration with Cursor
Add the server to your Cursor MCP configuration in ~/.cursor/mcp.json:
{
"mcpServers": {
"TalkToFigma": {
"command": "bunx",
"args": ["cursor-talk-to-figma-mcp@latest"]
}
}
}
WebSocket Server
Start the WebSocket server:
bun socket
Figma Plugin
- In Figma, go to Plugins > Development > New Plugin
- Choose "Link existing plugin"
- Select the
src/cursor_mcp_plugin/manifest.jsonfile - The plugin should now be available in your Figma development plugins
Windows + WSL Guide
- Install bun via powershell
powershell -c "irm bun.sh/install.ps1|iex"
- Uncomment the hostname
0.0.0.0insrc/socket.ts
// uncomment this to allow connections in windows wsl
hostname: "0.0.0.0",
- Start the websocket
bun socket
Usage
- Start the WebSocket server
- Install the MCP server in Cursor
- Open Figma and run the Cursor MCP Plugin
- Connect the plugin to the WebSocket server by joining a channel using
join_channel - Use Cursor to communicate with Figma using the MCP tools
Local Development Setup
To develop, update your mcp config to direct to your local directory.
{
"mcpServers": {
"TalkToFigma": {
"command": "bun",
"args": ["/path-to-repo/src/talk_to_figma_mcp/server.ts"]
}
}
}
MCP Tools
The MCP server provides the following tools for interacting with Figma:
Document & Selection
get_document_info- Get information about the current Figma documentget_selection- Get information about the current selectionread_my_design- Get detailed node information about the current selection without parametersget_node_info- Get detailed information about a specific nodeget_nodes_info- Get detailed information about multiple nodes by providing an array of node IDsset_focus- Set focus on a specific node by selecting it and scrolling viewport to itset_selections- Set selection to multiple nodes and scroll viewport to show them
Annotations
get_annotations- Get all annotations in the current document or specific nodeset_annotation- Create or update an annotation with markdown supportset_multiple_annotations- Batch create/update multiple annotations efficientlyscan_nodes_by_types- Scan for nodes with specific types (useful for finding annotation targets)
Prototyping & Connections
get_reactions- Get all prototype reactions from nodes with visual highlight animationset_default_connector- Set a copied FigJam connector as the default connector style for creating connections (must be set before creating connections)create_connections- Create FigJam connector lines between nodes, based on prototype flows or custom mapping
Creating Elements
create_rectangle- Create a new rectangle with position, size, and optional namecreate_frame- Create a new frame with position, size, and optional namecreate_section- Create a section to group related content on the canvascreate_text- Create a new text node with customizable font properties
Modifying text content
scan_text_nodes- Scan text nodes with intelligent chunking for large designsset_text_content- Set the text content of a single text nodeset_multiple_text_contents- Batch update multiple text nodes efficiently
Auto Layout & Spacing
set_layout_mode- Set the layout mode and wrap behavior of a frame (NONE, HORIZONTAL, VERTICAL)set_padding- Set padding values for an auto-layout frame (top, right, bottom, left)set_axis_align- Set primary and counter axis alignment for auto-layout framesset_layout_sizing- Set horizontal and vertical sizing modes for auto-layout frames (FIXED, HUG, FILL)set_item_spacing- Set distance between children in an auto-layout frame
Styling
set_fill_color- Set the fill color of a node (RGBA)set_stroke_color- Set the stroke color and weight of a nodeset_corner_radius- Set the corner radius of a node with optional per-corner controlset_image_fill- Fill a node with an image from a local file path, URL, or base64 data (FILL, FIT, CROP, or TILE)
Layout & Organization
move_node- Move a node to a new positionresize_node- Resize a node with new dimensionsdelete_node- Delete a nodedelete_multiple_nodes- Delete multiple nodes at once efficientlyclone_node- Create a copy of an existing node with optional position offsetrename_node- Rename a nodeset_parent- Move a node into a new parent (section, frame or group), preserving its absolute position by default
Components & Styles
get_styles- Get information about local stylesget_local_components- Get information about local componentscreate_component_instance- Create an instance of a componentget_instance_overrides- Extract override properties from a selected component instanceset_instance_overrides- Apply extracted overrides to target instances
Export & Advanced
export_node_as_image- Export a node as an image (PNG, JPG, SVG, or PDF) - limited support on image currently returning base64 as text
Connection Management
join_channel- Join a specific channel to communicate with Figma
MCP Prompts
The MCP server includes several helper prompts to guide you through complex design tasks:
design_strategy- Best practices for working with Figma designsread_design_strategy- Best practices for reading Figma designstext_replacement_strategy- Systematic approach for replacing text in Figma designsannotation_conversion_strategy- Strategy for converting manual annotations to Figma's native annotationsswap_overrides_instances- Strategy for transferring overrides between component instances in Figmareaction_to_connector_strategy- Strategy for converting Figma prototype reactions to connector lines using the output of 'get_reactions', and guiding the use 'create_connections' in sequence
Development
Building the Figma Plugin
Navigate to the Figma plugin directory:
cd src/cursor_mcp_pluginEdit code.js and ui.html
Best Practices
When working with the Figma MCP:
- Always join a channel before sending commands
- Get document overview using
get_document_infofirst - Check current selection with
get_selectionbefore modifications - Use appropriate creation tools based on needs:
create_framefor containerscreate_rectanglefor basic shapescreate_textfor text elements
- Verify changes using
get_node_info - Use component instances when possible for consistency
- Handle errors appropriately as all commands can throw exceptions
- For large designs:
- Use chunking parameters in
scan_text_nodes - Monitor progress through WebSocket updates
- Implement appropriate error handling
- Use chunking parameters in
- For text operations:
- Use batch operations when possible
- Consider structural relationships
- Verify changes with targeted exports
- For converting legacy annotations:
- Scan text nodes to identify numbered markers and descriptions
- Use
scan_nodes_by_typesto find UI elements that annotations refer to - Match markers with their target elements using path, name, or proximity
- Categorize annotations appropriately with
get_annotations - Create native annotations with
set_multiple_annotationsin batches - Verify all annotations are properly linked to their targets
- Delete legacy annotation nodes after successful conversion
- Visualize prototype noodles as FigJam connectors:
- Use
get_reactionsto extract prototype flows, - set a default connector with
set_default_connector, - and generate connector lines with
create_connectionsfor clear visual flow mapping.
FAQ
Common questions
Discussion
Questions & comments · 0
Sign In Sign in to leave a comment.