Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Direct Figma API interactions for design asset management. Fetch files and components, extract design tokens, export images, manage comments, and access version history.
.claude/skills/a5c-ai-figma-api/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | 40% | 0% |
| case-09 | ✗→✓ | ▲ Improved | 50% | 0% |
| case-23 | ✗→✓ | ▲ Improved | 155% | 0% |
| case-25 | ✗→✓ | ▲ Improved | 14% | 0% |
| case-26 | ✗→✓ | ▲ Improved | 29% | 0% |
You are figma-api - a specialized skill for direct Figma API interactions, enabling seamless design-to-code workflows and design asset management.
This skill enables AI-powered Figma integration including:
Retrieve Figma file data and components:
javascript// Fetch entire file const fileData = await figmaApi.getFile(fileKey); // Fetch specific nodes const nodes = await figmaApi.getFileNodes(fileKey, ['1:2', '1:3']); // Fetch component metadata const components = await figmaApi.getComponents(fileKey); // Fetch component sets (variants) const componentSets = await figmaApi.getComponentSets(fileKey);
Extract design tokens from Figma files:
json{ "colors": { "primary": { "50": { "value": "#E3F2FD", "type": "color" }, "100": { "value": "#BBDEFB", "type": "color" }, "500": { "value": "#2196F3", "type": "color" }, "900": { "value": "#0D47A1", "type": "color" } }, "semantic": { "success": { "value": "{colors.green.500}", "type": "color" }, "error": { "value": "{colors.red.500}", "type": "color" }, "warning": { "value": "{colors.yellow.500}", "type": "color" } } }, "typography": { "heading-1": { "fontFamily": { "value": "Inter", "type": "fontFamily" }, "fontSize": { "value": "48px", "type": "dimension" }, "fontWeight": { "value": "700", "type": "fontWeight" }, "lineHeight": { "value": "1.2", "type": "number" } } }, "spacing": { "xs": { "value": "4px", "type": "dimension" }, "sm": { "value": "8px", "type": "dimension" }, "md": { "value": "16px", "type": "dimension" }, "lg": { "value": "24px", "type": "dimension" }, "xl": { "value": "32px", "type": "dimension" } } }
Export images and assets at various scales:
javascript// Export specific nodes as PNG const images = await figmaApi.getImage(fileKey, { ids: ['1:2', '1:3'], format: 'png', scale: 2 }); // Export as SVG const svgImages = await figmaApi.getImage(fileKey, { ids: ['1:4'], format: 'svg', svg_include_id: true, svg_simplify_stroke: true }); // Export with fills rendered const renderedImages = await figmaApi.getImageFills(fileKey);
Manage design feedback and comments:
javascript// Get all comments const comments = await figmaApi.getComments(fileKey); // Post new comment const newComment = await figmaApi.postComment(fileKey, { message: 'Please review the button states', client_meta: { x: 100, y: 200 } }); // Reply to comment const reply = await figmaApi.postComment(fileKey, { message: 'Updated per feedback', comment_id: '123456' }); // Resolve comment await figmaApi.deleteComment(fileKey, commentId);
Access file version history:
javascript// Get version history const versions = await figmaApi.getVersions(fileKey); // Output: { "versions": [ { "id": "123456789", "created_at": "2026-01-24T10:30:00Z", "label": "v2.0 - Updated color system", "description": "Migrated to new brand colors", "user": { "id": "user_id", "handle": "designer", "img_url": "avatar.png" } } ] }
Extract styles from Figma:
javascript// Get all styles const styles = await figmaApi.getStyles(fileKey); // Extract color styles const colorStyles = styles.filter(s => s.style_type === 'FILL'); // Extract text styles const textStyles = styles.filter(s => s.style_type === 'TEXT'); // Extract effect styles (shadows, blurs) const effectStyles = styles.filter(s => s.style_type === 'EFFECT');
This skill can leverage the following MCP servers for enhanced capabilities:
| Server | Description | Installation | |--------|-------------|--------------| | Claude Talk to Figma MCP | Bidirectional Figma interaction for real-time design manipulation | GitHub | | Figma MCP Server (karthiks3000) | Claude MCP Server for working with Figma files | GitHub | | html.to.design MCP | Converts HTML directly into editable Figma designs | Docs |
| Endpoint | Method | Description | |----------|--------|-------------| | /v1/files/:key | GET | Get file data | | /v1/files/:key/nodes | GET | Get specific nodes | | /v1/files/:key/images | GET | Export images | | /v1/files/:key/comments | GET/POST | Manage comments | | /v1/files/:key/versions | GET | Get version history | | /v1/files/:key/components | GET | Get components |
bash# Using Personal Access Token curl -H "X-Figma-Token: YOUR_TOKEN" \ "https://api.figma.com/v1/files/FILE_KEY" # Using OAuth curl -H "Authorization: Bearer OAUTH_TOKEN" \ "https://api.figma.com/v1/files/FILE_KEY"
This skill integrates with the following processes:
component-library.js - Design-to-code component workflowsdesign-system.js - Design system synchronizationhifi-prototyping.js - High-fidelity prototype exportswireframing.js - Wireframe asset managementWhen executing operations, provide structured output:
json{ "operation": "extract-tokens", "fileKey": "abc123xyz", "status": "success", "tokens": { "colors": {}, "typography": {}, "spacing": {} }, "metadata": { "lastModified": "2026-01-24T10:30:00Z", "version": "123456789" }, "artifacts": ["tokens.json", "tokens.css"] }
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 29,241 | 13,922 | -52% | 1 | 1 | 0% | 3,109 | 4,361 | +40% | 0 | 0 | — |
case-02 | fail→fail | 20,111 | 15,393 | -23% | 1 | 1 | 0% | 764 | 4,361 | +471% | 0 | 0 | — |
case-03 | fail→fail | 18,002 | 10,273 | -43% | 1 | 1 | 0% | 466 | 3,243 | +596% | 0 | 0 | — |
case-04 | fail→fail | 15,374 | 13,749 | -11% | 1 | 1 | 0% | 1,843 | 3,633 | +97% | 0 | 0 | — |
case-05 | pass→pass | 16,597 | 14,331 | -14% | 1 | 1 | 0% | 2,032 | 3,889 | +91% | 0 | 0 | — |
case-06 | pass→pass | 16,630 | 16,651 | +0% | 1 | 1 | 0% | 2,198 | 3,851 | +75% | 0 | 0 | — |
case-07 | pass→pass | 10,088 | 8,210 | -19% | 1 | 1 | 0% | 1,023 | 2,727 | +167% | 0 | 0 | — |
case-08 | pass→pass | 10,445 | 9,898 | -5% | 1 | 1 | 0% | 1,037 | 3,011 | +190% | 0 | 0 | — |
case-09 | fail→pass | 20,560 | 13,441 | -35% | 1 | 1 | 0% | 2,441 | 3,652 | +50% | 0 | 0 | — |
case-10 | pass→pass | 17,308 | 12,682 | -27% | 1 | 1 | 0% | 2,147 | 3,589 | +67% | 0 | 0 | — |
case-11 | pass→pass | 8,460 | 9,826 | +16% | 1 | 1 | 0% | 586 | 2,889 | +393% | 0 | 0 | — |
case-12 | pass→pass | 11,547 | 8,862 | -23% | 1 | 1 | 0% | 1,304 | 2,919 | +124% | 0 | 0 | — |
case-13 | fail→fail | 14,647 | 12,410 | -15% | 1 | 1 | 0% | 2,003 | 3,564 | +78% | 0 | 0 | — |
case-14 | pass→pass | 13,126 | 12,005 | -9% | 1 | 1 | 0% | 1,361 | 3,366 | +147% | 0 | 0 | — |
case-15 | pass→pass | 15,708 | 14,579 | -7% | 1 | 1 | 0% | 2,039 | 3,892 | +91% | 0 | 0 | — |
case-16 | pass→pass | 26,872 | 10,615 | -60% | 1 | 1 | 0% | 1,515 | 3,086 | +104% | 0 | 0 | — |
case-17 | pass→pass | 18,130 | 13,429 | -26% | 1 | 1 | 0% | 2,139 | 3,682 | +72% | 0 | 0 | — |
case-18 | pass→pass | 9,901 | 8,083 | -18% | 1 | 1 | 0% | 755 | 2,546 | +237% | 0 | 0 | — |
case-19 | pass→pass | 13,800 | 10,696 | -22% | 1 | 1 | 0% | 1,669 | 3,091 | +85% | 0 | 0 | — |
case-20 | pass→pass | 13,411 | 8,055 | -40% | 1 | 1 | 0% | 1,409 | 2,723 | +93% | 0 | 0 | — |
case-21 | pass→pass | 11,436 | 9,990 | -13% | 1 | 1 | 0% | 1,093 | 2,982 | +173% | 0 | 0 | — |
case-22 | pass→pass | 15,440 | 3,789 | -75% | 1 | 1 | 0% | 2,046 | 2,819 | +38% | 0 | 0 | — |
case-23 | fail→pass | 6,260 | 10,504 | +68% | 1 | 1 | 0% | 1,061 | 2,701 | +155% | 0 | 0 | — |
case-24 | pass→pass | 14,872 | 11,620 | -22% | 1 | 1 | 0% | 1,575 | 3,237 | +106% | 0 | 0 | — |
case-25 | fail→pass | 37,233 | 4,300 | -88% | 1 | 1 | 0% | 2,451 | 2,788 | +14% | 0 | 0 | — |
case-26 | fail→pass | 13,959 | 8,926 | -36% | 1 | 1 | 0% | 2,170 | 2,790 | +29% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 26 cases were attempted, and 24 counted toward the lift figure. The other 2 produced results that are not comparable between the two arms, so they are excluded from the headline rather than averaged into it. The headline lift of +19 percentage points is the difference between those two pass rates over the 24 comparable cases.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.