figma-spec-mcp
Bridge Figma designs to any platform with structured specs for Unity, React, Flutter, SwiftUI, and more, including layout audit, design tokens, and accessibility checks.
README
figma-spec-mcp
<p align="center"> <img src="assets/banner.png" alt="figma-spec-mcp — Bridge Figma to Game Engines" width="720" /> </p>
Bridge Figma to any platform — Unity UGUI mappings built-in, structured output for React, Flutter, SwiftUI, and more. Layout audit, design tokens, accessibility checks, prototype flows, version diffs, and platform-ready specs — all through MCP.
Works with any MCP-compatible client: Claude Code, Claude Desktop, Cursor, VS Code + Copilot, Windsurf, Cline, Continue.dev, Zed.
Security note: Your Figma access token is passed as a tool argument. Never commit it to version control. Use environment variables or your AI client's secret management to supply it at runtime.
Platform Support
| Platform | What you get |
|---|---|
| Unity | map_to_unity — RectTransform data, layout groups, anchoring, UGUI component mapping |
| React | map_to_react — JSX tree, Tailwind/CSS classes, shadcn/MUI/Chakra/Radix component mapping, TypeScript props |
| React Native | map_to_react_native — View/Text/TouchableOpacity, StyleSheet, React Native Paper/NativeBase mapping |
| Flutter | map_to_flutter — Widget tree, Material/Cupertino components, ThemeData extraction |
| SwiftUI | map_to_swiftui — VStack/HStack/ZStack, modifier chains, Color asset extraction |
| Tailwind CSS | extract_design_tokens → Tailwind config export |
| Any platform | generate_implementation_contract → structured spec with scope, assets, states, and acceptance criteria |
Your AI agent reads the structured output and generates platform-specific code. No manual translation needed.
Quick Start
1. Get a Figma access token → figma.com/developers/api#access-tokens
2. Add to your MCP config (Claude Desktop, Cursor, VS Code, or any MCP client):
{
"mcpServers": {
"figma-spec-mcp": {
"command": "npx",
"args": ["-y", "figma-spec-mcp@beta"]
}
}
}
3. Restart your AI client and use the tools.
Your file key is in the Figma URL: figma.com/file/<FILE_KEY>/...
Why figma-spec-mcp?
Most Figma MCP tools forward raw API responses. figma-spec-mcp adds stable envelopes, focused derivations, and reusable engineering outputs:
- Deterministic JSON responses with a shared response envelope
- Built-in disk cache with freshness metadata on every result
- Source traceability for tokens, mappings, and extracted relationships
- Platform-ready outputs for Unity, codebase mapping, and image export workflows
Tools
inspect_layout— Inspects a Figma frame and returns hierarchy, layout structure, spacing, constraints, annotations, and basic accessibility warnings.extract_design_tokens— Extracts color, typography, and spacing tokens from a Figma file and exports them as CSS variables, Style Dictionary JSON, or Tailwind config.map_to_react— Maps a Figma frame to a React component tree with Tailwind/CSS classes, component library suggestions (shadcn, MUI, Chakra, Radix), asset hints, and TypeScript prop interfaces from variants.map_to_react_native— Maps a Figma frame to React Native components (View, Text, TouchableOpacity, TextInput) with StyleSheet output and React Native Paper/NativeBase library mapping.map_to_flutter— Maps a Figma frame to a Flutter Widget tree (Container, Column, Row, Stack) with Material/Cupertino component mapping and ThemeData extraction.map_to_swiftui— Maps a Figma frame to SwiftUI views (VStack, HStack, ZStack) with modifier chains and Color asset extraction.map_to_unity— Produces a Unity UGUI-oriented mapping with RectTransform data, layout groups, suggested components, notes, and warnings.resolve_components— Resolves instance nodes to their backing component definitions and returns source file and source node references.extract_flows— Extracts prototype transitions from a page or frame and returns directed flow connections plus a deterministic frame order.bridge_to_codebase— Scans a local project and maps Figma component names to likely implementation files using filename heuristics.diff_versions— Compares two Figma file versions and reports added, removed, and modified nodes.extract_variants— Reads a component set and returns structured variant metadata, parsed properties, dimensions, layout details, fills, and typography.export_images— Exports one or more Figma nodes as PNG, JPG, SVG, or PDF and returns the image URLs.audit_accessibility— Audits a frame for accessibility issues such as contrast, touch targets, font size, missing alt text, and color-only distinctions.simplify_context— Produces a token-efficient, LLM-oriented summary tree by collapsing wrappers, grouping repeated nodes, and truncating deep hierarchies.lint_handoff_readiness— Audits a frame for engineering handoff readiness: unnamed layers, absolute positioning soup, missing auto-layout, orphaned nodes, oversized images.generate_implementation_contract— Produces a structured implementation spec with scope, assets, states, interactions, dependencies, typography, colors, and acceptance criteria.extract_missing_states— Scans components for missing UI states (hover, pressed, disabled, loading, error, empty) against a standard expected-state set.flow_to_test_cases— Converts prototype flows into QA-ready test cases with navigation steps, expected outcomes, and flow coverage gaps.
Features
v0.1 — Core
inspect_layout,extract_design_tokens,map_to_unity- Disk cache with SHA-256 keying and 1h TTL
v0.2 — Intelligence
- Token name preservation from Figma styles
- Depth-limited chunking for large files
- Mixed/rich text runs extraction
- Annotation extraction, framework-aware hints (Unity, React, SwiftUI, Web)
v0.3 — Workflows
resolve_components— multi-file component traversalextract_flows— prototype flow graphbridge_to_codebase— Figma → repo file matchingdiff_versions— structured version diffextract_variants— component set batch extraction
v0.4 — Quality & DX
export_images— PNG/JPG/SVG/PDF exportaudit_accessibility— WCAG 2.1 contrast, touch targets, font sizesimplify_context— AI-optimized, token-efficient output- Tool registry pattern for easy contribution
- Rate limit handling (429 + Retry-After)
v0.5 — Handoff & QA
lint_handoff_readiness— design-to-code readiness audit with scoringgenerate_implementation_contract— structured implementation scope + acceptance criteriaextract_missing_states— component state coverage analysisflow_to_test_cases— prototype flows → QA test scenarios
v0.6 — Multi-Platform Mappers
map_to_react— Figma → React with Tailwind/CSS, shadcn/MUI/Chakra/Radix, TypeScript propsmap_to_react_native— Figma → React Native with StyleSheet, Paper/NativeBase component mappingmap_to_flutter— Figma → Flutter Widget tree with Material/Cupertino, ThemeData extractionmap_to_swiftui— Figma → SwiftUI with modifier chains, Color asset extraction- Normalized UI AST foundation — shared platform-agnostic tree powering all 4 platform mappers
Response Shape
All 19 tools return a consistent top-level envelope:
{
"schema_version": "0.1.0",
"source": { "file_key": "abc123", "node_id": "1:23" },
"freshness": {
"fresh": true,
"timestamp": "2026-03-26T10:00:00.000Z",
"ttl_ms": 3600000
},
"warnings": [],
"data": {}
}
Tool-specific results live in data, and most tools also include low-level cache metadata there.
Caching
Responses are cached to disk (default: $TMPDIR/figma-spec-mcp-cache/) by file key and request shape with a 1-hour TTL. Cache metadata is included in responses:
"cache": {
"cachedAt": "2026-03-26T10:00:00.000Z",
"expiresAt": "2026-03-26T11:00:00.000Z",
"fileVersion": "123456789",
"fresh": true
}
Development
git clone https://github.com/zaferdace/figma-spec-mcp
cd figma-spec-mcp
npm install
npm run build
node dist/index.js
Roadmap
- [x]
map_to_react— React component tree with Tailwind/CSS, component library mapping, TypeScript props - [x]
map_to_react_native— React Native with StyleSheet, Paper/NativeBase mapping - [x]
map_to_flutter— Flutter Widget tree with Material/Cupertino, ThemeData - [x]
map_to_swiftui— SwiftUI views with modifier chains, Color assets - [ ]
detect_design_drift— Compare Figma design against codebase implementation - [ ]
map_to_unreal_umg— Unreal Engine UMG widget mapping - [ ]
map_to_godot_control— Godot Control node mapping - [ ] Webhook-triggered spec generation
License
MIT © Zafer Dace
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
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.
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.