Build Figma

OrganizationPopular

cursor

build-figma

Use for "/build-figma", a figma.com/design URL with node-id, or "implement this Figma" into a web, desktop, or shared UI. Orchestrates Figma MCP intake, maps nodes to the repo's own design system, then a verify-this visual judge. Do not paste Figma Tailwind.

Overview

Publisher	cursor
Repository	`plugins`
Skill name	`build-figma`
Stars	9.2K
Forks	863
Bundled files	3
Links	Skill source Repository

Markdown instructions
A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.
Works with any LLM
AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.
3 bundled files
Scripts, templates, and references the model can read while it works. Files are read-only and never executed.
Open source
Published by cursor on GitHub. Read the source before you install it.

Installation

Install the Build Figma AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

Open the app and go to Plugins → Skills.
Choose "Install from GitHub".
Paste the skill folder URL below and confirm.
https://github.com/cursor/plugins/tree/main/dyl-stack/skills/build-figma
Enable the skill in any chat where you want it available.

Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills

git clone --depth 1 https://github.com/cursor/plugins.git /tmp/plugins
mkdir -p .claude/skills
cp -r /tmp/plugins/dyl-stack/skills/build-figma .claude/skills/build-figma

Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Build Figma in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Build Figma on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Build Figma is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

Loaded only when it is needed
The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.
Switch models mid-chat
Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Build Figma

Orchestrator for Figma → production UI. It does not replace the skills below. Read each when that phase starts; do not restate them here.

Requires figma (plugin and connected MCP) and cursor-team-kit. Check first with ../dyl-mode/references/requirements.md. Missing → stop and tell the user to run /add-plugin <name>.

Failure modes that shaped the gates: references/failure-lessons.md.

Delegate to (do not copy)

Phase	Read
Figma MCP mechanics	`figma-design-to-code` from the Figma plugin, before every `get_design_context`
Design-system inventory	references/design-system-discovery.md, then any repo-local design-system skill or rule it finds
Visual claim / verdict shape	`verify-this` from `cursor-team-kit` + references/visual-judge.md
Drive live UI	`control-ui` from `cursor-team-kit`, or the repo's own control skill for that surface

Gates


Build-figma progress:
- [ ] 0. Parse fileKey + nodeId (refuse file-only URLs)
- [ ] 1. Intake done before any product UI edit (see below)
- [ ] 2. Mapped to the repo's primitives/tokens only
- [ ] 3. Styling in the surface's own styling system
- [ ] 4. Visual judge VERIFIED per visual-judge.md (verify-this claim shape)

Gate 1 blocks coding. "Functional first, polish visuals later" is the main failure mode.

Workflow (thin)

0. Parse URL

figma.com/design/:fileKey/...?node-id=1-2 → fileKey, nodeId (1-2 → 1:2). Branch URLs use branchKey as fileKey. No node-id → stop.

1. Intake (blocking)

Follow figma-design-to-code. Call get_design_context and get_screenshot. If the response is sparse or too large, split to implementable child nodes. Download brand assets from MCP URLs; never redraw logos.

Run design-system discovery. Then write /tmp/build-figma/<slug>/intake.md:

Target surface, code path, and how you will drive it live.
Node tree.
Metric table. Figma px → token, or pinned value when no token matches.
Node → primitive map. Each Figma node to an existing component. Mark any node with no match as new, with why.

2. Implement

Use the surface's styling system and the mapped components. When Figma px disagree with a primitive's default size, pin the Figma metric in that styling system. Do not bend the primitive's defaults globally. Hard bans below.

3. Visual judge (blocking)

Specialize verify-this with references/visual-judge.md. Treat the Figma shot as baseline and the live capture as treatment. Match theme polarity first, then apply the rubric and write the verdict. After NOT VERIFIED, fix the failed rows, recapture, and apply the rubric again. Repeat until VERIFIED or INCONCLUSIVE. Stop and report the blocker after INCONCLUSIVE; do not claim Gate 4 complete.

4. Report

Intake path, map highlights, verdict + evidence paths, remaining gaps.

Hard bans

Paste Figma MCP Tailwind/React as product code.
Invent brand assets when Figma or the repo has them.
"Close enough" primitive for a distinct Figma structure.
Trust text/button/dialog size tokens without checking Figma px.
Claim visual done without the judge.
Edit product UI before intake.md exists.

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Build Figma AI skill do?

Why use Build Figma on TypingMind?

Because you install it once and use it with any model. Build Figma is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Build Figma in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/cursor/plugins/tree/main/dyl-stack/skills/build-figma. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Build Figma?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Build Figma?

As many as you like. As long as a model supports skills, you can use Build Figma with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Build Figma AI skill free?

It is published on GitHub by cursor. Check the repository for licensing terms. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Hyperframes

NousResearch

OrganizationPopular

Render MP4/WebM videos from HTML compositions.

Ip As Logo

Design minimal cute IP mascot marks readable at 32px.

Mono Color

Generate one- or two-ink editorial print poster images.

Pretext

Build creative browser demos with DOM-free text layout.

Overview

Installation

Install in TypingMind

Install in another agent

Use it in TypingMind

Works with any AI model

Skill instructions

Build Figma

Delegate to (do not copy)

Gates

Workflow (thin)

0. Parse URL

1. Intake (blocking)

2. Implement

3. Visual judge (blocking)

4. Report

Hard bans

Bundled files

Frequently asked questions

Related AI Skills

Set up your own AI workspace now