agentleFS
ek0212ek0212/agent-ready-repo-kit/skills/figma-to-ui/SKILL.md

figma-to-ui

Skill0 starsChanged 4 months agoMITView on GitHub

Implement an exact Figma frame, screen, or component in an existing frontend repository and refine it through browser comparison. Use when user provides a Figma selection link, says "turn this Figma into code", "implement this Figma design", "match this Figma frame", or asks to reproduce Figma UI in React, Next.js, Vue, Svelte, or HTML/CSS.

The file

We split this skill into the rules it gives an agent

8 rules. About 0 are common, 7 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈7

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Use for fidelity implementation from exact Figma node.

    line 11
  2. Distinctive:

    Use authenticated Figma connector.

    line 32
  3. Distinctive:

    Do not add Tailwind, component library, icon library, or styling system unless repository already uses it or user approves dependency.

    line 52
  4. Distinctive:

    Use supplied assets. Do not replace exact icons/images with emoji or unrelated library icons.

    line 57
  5. Distinctive:

    Run page. Capture at reference dimensions, representative mobile width, and each supplied breakpoint variant.

    line 65
  6. Distinctive:

    Do not claim pixel-perfect match without direct comparison.

    line 86
  7. Distinctive:

    Use direct simple sentences.

    line 90
  8. Distinctive:

    No mobile design: derive from repository breakpoints and label responsive assumptions.

    line 96

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Turn one exact Figma selection into repository-native responsive UI. Figma evidence and running page define acceptance.

Scope

Use for fidelity implementation from exact Figma node. Use frontend-design for open-ended visual direction without Figma source.

Requires authenticated Figma access, target repository, and browser or screenshot tool. User owns product behavior and design intent. Do not infer hidden nodes, assets, or interactions as facts.

Inputs

  • Exact Figma selection URL.
  • Target route, page, or component when known.
  • Required states and interactions.
  • Existing repository and run command.

Infer route, breakpoints, and component reuse only from repository and Figma evidence. State inference.

Workflow

Step 1: Pin Selection

Confirm URL resolves to intended node, not file root. Record node name, type, reference dimensions, variants, target entry point, and required states.

Step 2: Pull Evidence

Use authenticated Figma connector. Retrieve exact-node:

  • Screenshot and child hierarchy.
  • Auto-layout, constraints, grid, and dimensions.
  • Components, variants, variables, styles, type, spacing, radius, shadows, and effects.
  • Exportable images, SVGs, icons, and prototype links.

Screenshot is visual source. Structured values guide code. Inspect child nodes when parent omits required detail.

Step 3: Inspect Repository

Identify framework, package manager, routing, run/test commands, components, tokens, type, spacing, breakpoints, icon system, asset paths, and state patterns.

Map each Figma element to:

  • Existing component or token: reuse.
  • Durable missing variant: extend.
  • Design-specific element: create locally.
  • Exported visual: store through existing asset convention.

Do not add Tailwind, component library, icon library, or styling system unless repository already uses it or user approves dependency.

Step 4: Implement

  • Preserve routing, data, state, accessibility, and styling conventions.
  • Use supplied assets. Do not replace exact icons/images with emoji or unrelated library icons.
  • Match content, hierarchy, type, size, spacing, color, radius, borders, shadows, and alignment.
  • Implement visible interaction states and in-scope prototype behavior.
  • Build fluid layout between known frame sizes. Avoid coordinates valid only at reference viewport.
  • Preserve keyboard focus, semantic controls, alt text, and reduced motion.

Step 5: Compare

Run page. Capture at reference dimensions, representative mobile width, and each supplied breakpoint variant.

Correct in order:

  1. Structure, content, assets.
  2. Container, grid, alignment, responsive behavior.
  3. Component dimensions and spacing.
  4. Type and icon size.
  5. Color, border, radius, shadow, effects.
  6. Hover, focus, pressed, loading, transition behavior.

Reload after material changes. Stop only when visible differences are resolved or named.

Verify

  • Reference layout closely matches screenshot.
  • Mobile has no clipping, overflow, or unreadable density.
  • Required interactions work.
  • Narrow relevant tests, type check, lint, or build pass.
  • Remaining differences tied to named constraint.

Do not claim pixel-perfect match without direct comparison.

Writing

Use direct simple sentences. Skip rhetorical flourishes. Avoid X-not-Y contrasts, self-answering rhetorical questions, em dashes, alliterative three-part phrases, vague inspirational pivots, unsourced claims, and unverified quotes.

Fallbacks

  • Figma auth failure: request accessible selection URL or exported screenshot.
  • Missing asset/font: use approved repository fallback and name difference.
  • No mobile design: derive from repository breakpoints and label responsive assumptions.
  • Cannot run browser: implement static pass, state visual verification incomplete.

Output

Report files changed, route/component, checks run, viewports compared, assumptions, and remaining differences.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.