Skip to content
SkillModel-led2,488 tokens in use

stitch::generate-design

Generates Stitch screens from text or images, edits existing screens and creates design variants, with a prompt enhancement step first.

10,975 stars on the repository Apache License 2.0google-labs-code/stitch-skills · plugins/stitch-design/skills/generate-design
npx skills add google-labs-code/stitch-skills
View source

Install command as written in the repository's README

Skill label

measured from its files

What installing stitch::generate-design costs

54tokens idle
2,488tokens in use
4,070tokens, up to

Idle is what sits in context while the skill is installed. In use is its instructions, loaded when it runs. Up to adds the files its instructions link. Counts are approximate, made offline with cl100k_base.

Model dependence Model-led

Script-ledMixedModel-led

Mostly instructions, so how well it works depends on the model that follows them.

This describes the skill's structure, how much of it is scripts. It does not say how well it works.

Touches

shelldeclared and detectednetworkdetectedwrites filesdeclared

Declared tags come from the skill's frontmatter; detected ones from its scripts, its code blocks and the commands it names in inline code. This is detected, not audited.

How it scores

Quality

How well the skill is written, counted from its own files: its frontmatter, its steps, examples, stated output and constraints, its size and whether every file it ships is one it names.How we score it →
Excellent
  • Frontmatter4/15
  • Structure25/25
  • Examples15/15
  • Output15/15
  • Size13/15
  • Resources15/15

Security

Starts at 100 and loses points for each thing in the skill's files that a reader should look at first.How we score it →
Excellent

No findings.

Repository health

Pre-computed score based on development activity, maintenance, community, maturity, and trend momentum.How we score it →
Good
  • Development activity80
  • Maintenance52
  • Community68
  • Maturity32
  • Momentum40

This skill drives the Stitch MCP tools for design work. It improves each prompt with UI and UX terminology, then generates screens from text, recreates screens from uploaded images, edits screens with targeted prompts and produces variants.

Generated HTML and screenshots are downloaded to .stitch/designs, and the metadata file is kept current. It leaves colors and fonts to the project’s design system.

What You Get

  • Screens created in a Stitch project from a structured layout prompt
  • Edited screens that follow one focused change at a time
  • Up to five variants with a chosen creative range and aspects
  • Downloaded HTML and screenshots under .stitch/designs
  • An updated .stitch/metadata.json after edits

Common Use Cases

  • Creating a mobile browse tab from a short text idea
  • Recreating a dashboard mockup image as an editable Stitch screen
  • Exploring layout and color scheme variants of a home screen

What Makes It Unique

A prompt enhancement pipeline maps vague wording to professional UI terms from bundled reference files. The skill keeps colors, fonts and roundness out of generation prompts when a design system exists, which avoids conflicts with project-level tokens.

Under The Hood

The skill calls Stitch MCP tools such as list_projects, generate_screen_from_text, edit_screens and generate_variants. It downloads output files with curl through the shell and writes them under .stitch/designs. It delegates image upload to upload-to-stitch and design system setup to manage-design-system.

Join founders building with open source

Opinionated takes, migration guides, cost-saving tips, and insights from the open source ecosystem.

Subscribe on SubstackJoin 750+ subscribers