Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.uspec.design/llms.txt

Use this file to discover all available pages before exploring further.

uSpec 2.0 is here. Added new skill create-component-md. Bring your pixel-perfect Figma component to your favorite LLM. Read the article about Component.md →
uSpec: design system documentation for Figma components
uSpec turns your Figma components into structured documentation from your AI agent. Pick a component, a skill analyzes it, and the output lives either in Figma as annotation frames or on disk as a single .md spec you can take anywhere. Works with Cursor, Claude Code, and Codex.

​
Start here: Component Markdown

Your component. One .md file. Anywhere.

New in v2.0. Your design system source of truth, in one markdown file per component. Covers API, structure, color, and screen-reader behavior. An artifact that LLM tools can build from and humans can query.

​
Render into Figma

The original uSpec skills render annotation frames directly into your Figma file, next to the component.

Component Anatomy

Numbered markers and attribute tables for every element

Component API

Properties, values, defaults, and configuration examples

Component Properties

Variant axes, boolean toggles, and variable mode exhibits

Component Color

Design token mapping for every element and state

Component Structure

Dimensions, spacing, and padding across density and size variants

Screen Reader

VoiceOver, TalkBack, and ARIA accessibility specs

Motion

Animation timeline bars and easing details from After Effects

​
How it works

uSpec supports two rendering paths from the same source component:
1

Figma-native path

Point the skill at a Figma link. The agent uses the Figma MCP to pull structure, variables, and styles from your file, then renders annotation frames directly in Figma using templates you linked during setup.
2

Component Markdown path

Run the uSpec Extract Figma plugin to produce a _base.json capture of your component, then run create-component-md against the JSON. The skill runs four parallel interpretation agents (API, structure, color, voice) and writes a single .md specification to disk.

​
Get started

Setup

Set up your AI agent, Figma MCP, and your template library.

Create your first spec

Prompt format, available skills, and tips for better output.