Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Generate UI designs and frontend code with Google Stitch via MCP. Use when asked to create screens, mockups, UI designs, or generate frontend code from text descriptions. Supports desktop, mobile, and tablet layouts.
.claude/skills/dicklesworthstone-stitch/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | -72% | 0% |
| case-03 | ✗→✓ | ▲ Improved | 269% | 0% |
| case-07 | ✗→✓ | ▲ Improved | 76% | 0% |
| case-08 | ✗→✓ | ▲ Improved | 23% | 0% |
| case-09 | ✗→✓ | ▲ Improved | 65% | 0% |
Generate production-ready UI designs and frontend code using Google Stitch via MCP.
Verify stitch is configured:
bashmcporter list stitch
If missing, configure it:
bash# 1. Authenticate with Google Cloud gcloud auth application-default login # 2. Enable Stitch API gcloud beta services mcp enable stitch.googleapis.com --project=YOUR_PROJECT_ID # 3. Add to mcporter mcporter config add stitch --stdio "npx -y stitch-mcp" --env GOOGLE_CLOUD_PROJECT=YOUR_PROJECT_ID
| Tool | Purpose | |------|---------| | list_projects | List all projects | | create_project | Create a project | | get_project | Get project details | | list_screens | List screens in a project | | get_screen | Get screen details | | generate_screen_from_text | Generate a screen from a prompt | | fetch_screen_code | Download HTML/CSS code | | fetch_screen_image | Download preview image |
bashmcporter call stitch.list_projects mcporter call stitch.list_projects filter="view=owned" mcporter call stitch.list_projects filter="view=shared"
bashmcporter call stitch.create_project title="My App Design"
bashmcporter call stitch.generate_screen_from_text \ projectId="PROJECT_ID" \ prompt="A modern dashboard with sidebar navigation, user stats cards, and a data table" \ deviceType="DESKTOP" \ modelId="GEMINI_3_PRO"
Parameters:
projectId (required): Numeric project ID (e.g., 4651986810686364080)prompt (required): UI descriptiondeviceType: MOBILE (default), DESKTOP, TABLET, or AGNOSTICmodelId: GEMINI_3_FLASH (default, 30–60s) or GEMINI_3_PRO (1–3min, higher quality)Generation takes 1–3 minutes. Do not retry.
bashmcporter call stitch.list_screens projectId="PROJECT_ID"
bashmcporter call stitch.get_screen projectId="PROJECT_ID" screenId="SCREEN_ID"
bashmcporter call stitch.fetch_screen_code projectId="PROJECT_ID" screenId="SCREEN_ID"
Returns self-contained HTML with Tailwind CSS.
bashmcporter call stitch.fetch_screen_image projectId="PROJECT_ID" screenId="SCREEN_ID"
Saves PNG to current directory; returns base64 data.
A settings page with a left sidebar showing navigation categories
and a main content area with toggle switches for notifications,
privacy settings, and account preferences. Dark mode.An e-commerce product listing page with a search bar, filter sidebar
on the left, and a 3-column grid of product cards showing image,
title, price, and add-to-cart button.A nice websiteSettings page| Type | Resolution | Best For | |------|------------|----------| | DESKTOP | 1280×1024 | Full layouts, sidebars, wide tables | | MOBILE | 390×844 | Single column, bottom nav, touch targets | | TABLET | 768×1024 | Hybrid layouts | | AGNOSTIC | Varies | Responsive components |
fetch_screen_codeReference existing designs in subsequent prompts:
Following the same design system as the dashboard (dark mode,
Inter font, blue accent #137fec, rounded-lg corners),
create a user profile page with...Returns:
screenId: Generated screen IDoutput_components: Feedback or suggestionsIf output_components contains suggestions (e.g., "Yes, make them all"), present them to the user. If accepted, call generate_screen_from_text again with the suggestion as the prompt.
Generated HTML includes:
| Error | Cause | Fix | |-------|-------|-----| | HTTP 401 | Expired auth | gcloud auth application-default login | | HTTP 403 | API disabled | gcloud beta services mcp enable stitch.googleapis.com | | Timeout | Slow generation | Wait; check status with get_screen | | Project ID not found | Missing env var | Set GOOGLE_CLOUD_PROJECT |
bash# Create project mcporter call stitch.create_project title="SaaS Dashboard" # Generate dashboard mcporter call stitch.generate_screen_from_text \ projectId="NEW_PROJECT_ID" \ prompt="A SaaS analytics dashboard: header with logo and user avatar, left sidebar with nav (Dashboard, Analytics, Users, Settings), main area with 4 stat cards (Revenue, Users, Orders, Growth) and a line chart" \ deviceType="DESKTOP" \ modelId="GEMINI_3_PRO" # List screens mcporter call stitch.list_screens projectId="NEW_PROJECT_ID" # Download code mcporter call stitch.fetch_screen_code projectId="NEW_PROJECT_ID" screenId="SCREEN_ID"
bashmcporter call stitch.generate_screen_from_text \ projectId="PROJECT_ID" \ prompt="Mobile login screen: app logo at top, email and password fields, 'Forgot password?' link, primary 'Sign In' button, 'Don't have an account? Sign up' at bottom" \ deviceType="MOBILE" \ modelId="GEMINI_3_FLASH"
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-04 | pass→pass | 34,408 | 27,247 | -21% | 1 | 1 | 0% | 8,225 | 8,402 | +2% | 0 | 0 | — |
case-06 | pass→pass | 9,328 | 6,834 | -27% | 1 | 1 | 0% | 1,681 | 3,022 | +80% | 0 | 0 | — |
case-05 | pass→pass | 6,560 | 3,561 | -46% | 1 | 1 | 0% | 1,135 | 2,190 | +93% | 0 | 0 | — |
case-01 | fail→pass | 38,212 | 6,478 | -83% | 1 | 1 | 0% | 8,264 | 2,323 | -72% | 0 | 0 | — |
case-02 | fail→fail | 10,541 | 8,987 | -15% | 1 | 1 | 0% | 809 | 2,188 | +170% | 0 | 0 | — |
case-03 | fail→pass | 5,120 | 14,115 | +176% | 1 | 1 | 0% | 727 | 2,682 | +269% | 0 | 0 | — |
case-07 | fail→pass | 6,944 | 2,356 | -66% | 1 | 1 | 0% | 1,207 | 2,128 | +76% | 0 | 0 | — |
case-08 | fail→pass | 8,312 | 2,542 | -69% | 1 | 1 | 0% | 1,651 | 2,026 | +23% | 0 | 0 | — |
case-09 | fail→pass | 15,879 | 4,836 | -70% | 1 | 1 | 0% | 1,465 | 2,418 | +65% | 0 | 0 | — |
case-10 | fail→pass | 4,746 | 2,672 | -44% | 1 | 1 | 0% | 867 | 2,055 | +137% | 0 | 0 | — |
case-11 | fail→pass | 14,745 | 3,309 | -78% | 1 | 1 | 0% | 2,324 | 2,202 | -5% | 0 | 0 | — |
case-12 | fail→pass | 8,440 | 3,512 | -58% | 1 | 1 | 0% | 1,417 | 2,276 | +61% | 0 | 0 | — |
case-13 | fail→pass | 12,834 | 2,905 | -77% | 1 | 1 | 0% | 1,222 | 2,170 | +78% | 0 | 0 | — |
case-14 | fail→pass | 10,526 | 2,995 | -72% | 1 | 1 | 0% | 1,875 | 2,245 | +20% | 0 | 0 | — |
case-15 | fail→pass | 12,576 | 2,471 | -80% | 1 | 1 | 0% | 1,443 | 2,123 | +47% | 0 | 0 | — |
case-16 | fail→pass | 5,592 | 2,456 | -56% | 1 | 1 | 0% | 974 | 2,023 | +108% | 0 | 0 | — |
case-17 | fail→pass | 7,230 | 1,531 | -79% | 1 | 1 | 0% | 1,266 | 1,919 | +52% | 0 | 0 | — |
case-18 | fail→pass | 8,394 | 2,387 | -72% | 1 | 1 | 0% | 1,453 | 2,107 | +45% | 0 | 0 | — |
case-19 | pass→pass | 12,878 | 6,054 | -53% | 1 | 1 | 0% | 1,988 | 2,694 | +36% | 0 | 0 | — |
case-20 | fail→pass | 6,993 | 2,046 | -71% | 1 | 1 | 0% | 1,134 | 1,985 | +75% | 0 | 0 | — |
case-21 | fail→pass | 11,236 | 4,009 | -64% | 1 | 1 | 0% | 1,921 | 2,378 | +24% | 0 | 0 | — |
case-22 | pass→pass | 12,378 | 8,786 | -29% | 1 | 1 | 0% | 2,198 | 3,161 | +44% | 0 | 0 | — |
case-23 | pass→pass | 12,226 | 7,433 | -39% | 1 | 1 | 0% | 1,943 | 2,944 | +52% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 23 cases were attempted, and 22 counted toward the lift figure. The other 1 produced results that are not comparable between the two arms, so they are excluded from the headline rather than averaged into it. The headline lift of +70 percentage points is the difference between those two pass rates over the 22 comparable cases.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.