Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Professional Figma design analysis and asset export. Use for extracting design data, exporting assets in multiple formats, auditing accessibility compliance, analyzing design systems, and generating comprehensive design documentation. Read-only analysis of Figma files with powerful export and reporting capabilities.
.claude/skills/sundial-org-figma/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-13 | ✗→✓ | ▲ Improved | 105% | 0% |
| case-01 | ✗→✓ | ▲ Improved | -6% | 0% |
| case-02 | ✗→✓ | ▲ Improved | 379% | 0% |
| case-03 | ✗→✓ | ▲ Improved | 138% | 0% |
| case-04 | ✗→✓ | ▲ Improved | 106% | 0% |
Professional-grade Figma integration for design system analysis, asset export, and comprehensive design auditing.
Example usage:
Example usage:
Example usage:
Example usage:
bash# Set your Figma access token export FIGMA_ACCESS_TOKEN="your-token-here" # Or store in .env file echo "FIGMA_ACCESS_TOKEN=your-token" >> .env
bash# Get file information and structure python scripts/figma_client.py get-file "your-file-key" # Export frames as images python scripts/export_manager.py export-frames "file-key" --formats png,svg # Analyze design system consistency python scripts/style_auditor.py audit-file "file-key" --generate-html # Check accessibility compliance python scripts/accessibility_checker.py "file-key" --level AA --format html
figma_client.py - Complete Figma API wrapper with all REST endpointsexport_manager.py - Professional asset export with multiple formats and scalesstyle_auditor.py - Design system analysis and brand consistency checkingaccessibility_checker.py - Comprehensive WCAG compliance validation and reportingfigma-api-reference.md - Complete API documentation and examplesdesign-patterns.md - UI patterns and component best practicesaccessibility-guidelines.md - WCAG compliance requirementsexport-formats.md - Asset export options and specificationstemplates/design-system/ - Pre-built component library templatestemplates/brand-kits/ - Standard brand guideline structurestemplates/wireframes/ - Common layout patterns and flowsbash# Generate design tokens for CSS python scripts/export_manager.py export-tokens "file-key" --format css # Create component documentation python scripts/figma_client.py document-components "file-key" --output docs/
bash# Audit brand compliance in designs python scripts/style_auditor.py audit-file "file-key" --brand-colors "#FF0000,#00FF00,#0000FF" # Extract current brand colors for analysis python scripts/figma_client.py extract-colors "file-key" --output brand-colors.json
bash# Generate client presentation assets python scripts/export_manager.py client-package "file-key" --template presentation # Create development handoff assets python scripts/export_manager.py dev-handoff "file-key" --include-specs
This skill provides read-only access to Figma files through the REST API. It can:
For file modifications, you would need to develop a Figma plugin using the Plugin API.
Built-in rate limiting and retry logic to handle Figma's API constraints gracefully.
Comprehensive error handling with detailed logging and recovery suggestions.
Export assets in PNG, SVG, PDF, and WEBP with platform-specific sizing.
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-13 | fail→pass | 5,767 | 2,637 | -54% | 1 | 1 | 0% | 913 | 1,871 | +105% | 0 | 0 | — |
case-01 | fail→pass | 17,187 | 11,555 | -33% | 1 | 1 | 0% | 3,532 | 3,303 | -6% | 0 | 0 | — |
case-02 | fail→pass | 4,778 | 14,052 | +194% | 1 | 1 | 0% | 674 | 3,226 | +379% | 0 | 0 | — |
case-03 | fail→pass | 7,657 | 8,753 | +14% | 1 | 1 | 0% | 1,404 | 3,341 | +138% | 0 | 0 | — |
case-04 | fail→pass | 7,472 | 6,207 | -17% | 1 | 1 | 0% | 1,208 | 2,483 | +106% | 0 | 0 | — |
case-05 | fail→pass | 6,226 | 7,461 | +20% | 1 | 1 | 0% | 1,063 | 2,800 | +163% | 0 | 0 | — |
case-06 | fail→pass | 8,616 | 7,897 | -8% | 1 | 1 | 0% | 1,355 | 2,688 | +98% | 0 | 0 | — |
case-07 | fail→pass | 4,375 | 2,052 | -53% | 1 | 1 | 0% | 681 | 1,760 | +158% | 0 | 0 | — |
case-08 | fail→pass | 7,811 | 3,998 | -49% | 1 | 1 | 0% | 1,173 | 2,143 | +83% | 0 | 0 | — |
case-09 | fail→pass | 8,560 | 3,048 | -64% | 1 | 1 | 0% | 1,439 | 1,986 | +38% | 0 | 0 | — |
case-10 | fail→pass | 8,960 | 2,897 | -68% | 1 | 1 | 0% | 1,460 | 1,959 | +34% | 0 | 0 | — |
case-11 | fail→pass | 11,599 | 2,588 | -78% | 1 | 1 | 0% | 1,860 | 1,875 | +1% | 0 | 0 | — |
case-12 | fail→pass | 7,890 | 2,234 | -72% | 1 | 1 | 0% | 1,197 | 1,841 | +54% | 0 | 0 | — |
case-14 | fail→pass | 4,699 | 2,643 | -44% | 1 | 1 | 0% | 749 | 1,871 | +150% | 0 | 0 | — |
case-15 | pass→pass | 10,053 | 4,355 | -57% | 1 | 1 | 0% | 1,561 | 2,172 | +39% | 0 | 0 | — |
case-16 | fail→fail | 17,110 | 11,413 | -33% | 1 | 1 | 0% | 2,608 | 3,380 | +30% | 0 | 0 | — |
case-17 | fail→pass | 2,231 | 3,813 | +71% | 1 | 1 | 0% | 299 | 2,102 | +603% | 0 | 0 | — |
case-18 | fail→pass | 12,615 | 4,845 | -62% | 1 | 1 | 0% | 2,335 | 2,398 | +3% | 0 | 0 | — |
case-19 | pass→pass | 14,377 | 5,574 | -61% | 1 | 1 | 0% | 2,375 | 2,415 | +2% | 0 | 0 | — |
case-20 | pass→pass | 2,532 | 2,650 | +5% | 1 | 1 | 0% | 412 | 1,889 | +358% | 0 | 0 | — |
case-21 | fail→pass | 11,390 | 2,232 | -80% | 1 | 1 | 0% | 1,809 | 1,808 | -0% | 0 | 0 | — |
case-22 | fail→pass | 9,363 | 1,926 | -79% | 1 | 1 | 0% | 1,589 | 1,759 | +11% | 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. 22 cases were attempted. The headline lift of +82 percentage points is the difference between those two pass rates over the 22 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.