Review and verify specs on Figma’s canvas

Use the Specs 2 plugin to create specs on the canvas — styled your way and tuned to your library’s conventions.

Generate canvas output Add anatomy, props, layout, styling, modes and data Match naming conventions Detect your subcomponents, glyphs, and examples Custom styling Style color, space and text to your preferences Extract data to LLMs Copy structured spec data of a library

Run commands in LLM chats, agents, and pipelines

Generate – and regenerate – an entire library in a minute that feeds specs, code and more data into your agentic workflow

Generate library specs Process 50 components in a minute for your pipeline Set up an environment One-time setup for sources, settings and scanning

A system to synchronize your design intent

When every surface from prototype to production reads from one record, change stops being a handoff and starts to become sync.

Computation over inference

Don’t start agents prematurely. Extract from Figma faster, cheaper, and more reliably - over and over.

	Specs ecosystem	Agentic Figma extraction
Speed	~1 sec (CLI) / ~10 sec (Plugin) per component	~5–10 minutes per component
AI cost	Scripts require 0 AI tokens	25/50/100,000+ tokens per component
Repeatability	Deterministic, mechanical	Inference errors, gaps, and overconfidence
Output	Schema-valid YAML/MD, ideal for testing & versioning	Unstructured Markdown means unpredictable verification
Platforms	Schema-based, x-platform	Guessing your intents

Every layer and property accounted for

A single spec records everything in Figma about a component, deterministic and compact.

Anatomy Every element by type, structured consistently across variants

Props All properties with every explicit per-variant difference detected

Styling Token references, named styles, typography, and effects

Layout Element hierarchy and flex vs. absolute positioning per variant

Bindings Prop bindings driving content, instances, and conditionals

Subcomponents Subcomponents discovered from your naming conventions

Slots Where nested content composes and its per-variant defaults

Ready-made examples Pre-configured usage examples detected straight from your file

Assets Icon glyphs precisely distinguished from nested instances

Robust at the core, extensible at the edges

Blend determinism with adaptability to generate specs based on how your system works.

Variant Layering Compute layered styling like CSS rules

Variant Depth Control how many variant dimensions expand

Data Layout Shape how hierarchy appears in output

Consolidating Props Merge a toggle and its content into one prop

Code-Only Props Pull non-visual props from hidden layers

Instance Examples Detect ready-made, pre-configured usages

Default Slot Content Reuse default slot content as examples

Slot Constraints Set quantity and content-type slot rules

Subcomponent Scoping Discover, match, and exclude subcomponents

Invalid Variant Combinations Mark unsupported variant combinations

Key Formatting Normalize property and element names

Layout Positioning Map Figma constraints to positioning