templates/skills/modules/figma/SKILL.md
Use MCP Figma for design system integration, asset export, and design-to-code workflows.
npx skillsauth add hivellm/rulebook FigmaInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
CRITICAL: Use MCP Figma for design system integration, asset export, and design-to-code workflows.
// Get file
figma.getFile({ file_key: 'file-key' })
// Get file nodes
figma.getFileNodes({
file_key: 'file-key',
ids: ['node-id-1', 'node-id-2']
})
// Get file versions
figma.getFileVersions({ file_key: 'file-key' })
// Export images
figma.getImage({
file_key: 'file-key',
ids: 'node-id',
format: 'png', // png, jpg, svg, pdf
scale: 2 // @2x resolution
})
// Export multiple
figma.getImage({
file_key: 'file-key',
ids: 'node-1,node-2,node-3',
format: 'svg',
svg_outline_text: true
})
// Get components
figma.getFileComponents({ file_key: 'file-key' })
// Get component sets
figma.getFileComponentSets({ file_key: 'file-key' })
// Get team components
figma.getTeamComponents({ team_id: 'team-id' })
// Get file styles
figma.getFileStyles({ file_key: 'file-key' })
// Get team styles
figma.getTeamStyles({ team_id: 'team-id' })
// Get comments
figma.getComments({ file_key: 'file-key' })
// Post comment
figma.postComment({
file_key: 'file-key',
message: 'Approved for development',
comment_id: 'parent-comment-id' // For replies
})
// Export design tokens from Figma styles
const { data: { styles } } = await figma.getFileStyles({ file_key: fileKey })
const tokens = {
colors: {},
typography: {},
spacing: {}
}
for (const style of Object.values(styles)) {
if (style.style_type === 'FILL') {
tokens.colors[style.name] = extractColor(style)
} else if (style.style_type === 'TEXT') {
tokens.typography[style.name] = extractTextStyle(style)
}
}
// Write to tokens.json
fs.writeFileSync('tokens.json', JSON.stringify(tokens, null, 2))
// Sync Figma components to code
const { data: components } = await figma.getFileComponents({ file_key: fileKey })
for (const component of Object.values(components)) {
// Export component as SVG
const { data: images } = await figma.getImage({
file_key: fileKey,
ids: component.node_id,
format: 'svg'
})
// Save to assets folder
const svg = await fetch(images[component.node_id]).then(r => r.text())
fs.writeFileSync(`assets/icons/${component.name}.svg`, svg)
}
// Generate screenshots for documentation
const screens = ['home-screen', 'login-screen', 'dashboard']
for (const screenId of screens) {
const { data: images } = await figma.getImage({
file_key: fileKey,
ids: screenId,
format: 'png',
scale: 2
})
const imageUrl = images[screenId]
const response = await fetch(imageUrl)
const buffer = await response.buffer()
fs.writeFileSync(`docs/screenshots/${screenId}.png`, buffer)
}
// Check for new comments
const { data: comments } = await figma.getComments({ file_key: fileKey })
const unresolved = comments.filter(c => !c.resolved_at)
if (unresolved.length > 0) {
console.log(`${unresolved.length} unresolved design comments`)
// Create Jira issues for unresolved comments
for (const comment of unresolved) {
await jira.issues.createIssue({
fields: {
project: { key: 'DESIGN' },
summary: `Design feedback: ${comment.message.substring(0, 50)}`,
description: comment.message,
issuetype: { name: 'Task' }
}
})
}
}
// Generate component documentation
const { data: components } = await figma.getFileComponents({ file_key: fileKey })
let markdown = '# Design System Components\n\n'
for (const [id, component] of Object.entries(components)) {
// Export thumbnail
const { data: images } = await figma.getImage({
file_key: fileKey,
ids: id,
format: 'png',
scale: 1
})
markdown += `## ${component.name}\n\n`
markdown += `\n\n`
markdown += `**Description:** ${component.description || 'No description'}\n\n`
}
fs.writeFileSync('docs/design-system.md', markdown)
✅ DO:
❌ DON'T:
{
"mcpServers": {
"figma": {
"command": "npx",
"args": ["-y", "@modelcontextprotocol/server-figma"],
"env": {
"FIGMA_ACCESS_TOKEN": "your-personal-access-token"
}
}
}
}
Setup:
# Export icons on every design update
figma-export --file-key=$FIGMA_FILE --format=svg --output=src/assets/icons
# Optimize SVGs
svgo --folder src/assets/icons
# Commit if changes detected
git diff --quiet src/assets/icons || git commit -m "chore: Update design assets"
// 1. Detect design changes
const currentVersion = await figma.getFile({ file_key: fileKey })
const lastVersion = loadLastProcessedVersion()
if (currentVersion.version !== lastVersion) {
// 2. Export updated components
await exportComponents(fileKey)
// 3. Generate code
await generateComponentCode()
// 4. Run tests
await runVisualRegressionTests()
// 5. Create PR if tests pass
if (testsPass) {
await createPullRequest('Update components from Figma')
}
}
<!-- FIGMA:END -->research
Author a rulebook task spec interactively — research, draft, ask the user clarifying questions, confirm, then create the tasks in rulebook ready for /rulebook-driver. Use when the user wants to plan/spec a feature before implementing.
development
Behavioral guidelines to reduce common LLM coding mistakes — overcomplication, sloppy refactors, hidden assumptions, weak goals. Use when writing, reviewing, or refactoring code. Auto-applies; invoke explicitly via /karpathy-guidelines or 'follow karpathy discipline'.
data-ai
Autonomous AI agent loop for iterative task implementation (@hivehub/rulebook ralph)
data-ai
Use SQL Server for enterprise relational data storage with advanced features, high availability, and Windows integration.