React + Vite Dashboard SkillSAFE
A library of Agent Skills designed to work with the Stitch MCP server. Each skill follows the Agent Skills open standard, for compatibility with coding agents such as Antigravity, Gemini CLI, Claude Code, Cursor.
Overview
From the repository's own README, as read at the audited commit. Badges and raw HTML are left out.
Convert Stitch screens into production React dashboards with Vite, TanStack Query, and optional Web3 reads.
Install
npx skills add google-labs-code/stitch-skills --skill react-vite-dashboard --global
Example prompts
Convert the Stitch dashboard screen into React + Vite components using our DESIGN.md tokens
Wire the portfolio summary cards with TanStack Query and ethers read-only calls
Related skills
design-md— author and lint DESIGN.md before building UIshadcn-ui— add accessible component primitives on top of generated layouts
32484555efabOBSERVED · 2026-10-22Install
Commands as the repository documents them. They are shown, not run.
npx skills add google-labs-code/stitch-skills --skill react-vite-dashboard --global
What it tells the agent
The instruction file, verbatim from the audited commit — this is the text the model reads, and the surface the audit's instruction layer examines. Quoted here so you can judge it without cloning anything.
---
name: react-vite-dashboard
description: Convert Stitch designs into production React + Vite dashboards with TanStack Query, accessible tokens from DESIGN.md, and Web3-ready patterns (ethers/viem).
allowed-tools:
- "stitch*:*"
- "Read"
- "Write"
- "Bash"
- "web_fetch"
---
# Stitch to React + Vite Dashboard
You are a frontend engineer building **data-dense dashboards** from Stitch screens. Target stack: **React 18**, **Vite**, **TypeScript**, **TanStack Query**, **React Router**, and optional **ethers v6** or **viem** for on-chain reads.
## Prerequisites
- Stitch MCP configured ([setup guide](https://stitch.withgoogle.com/docs/mcp/setup/))
- A project `DESIGN.md` (see the `design-md` skill) for token fidelity
- Vite + React + TypeScript scaffold (`npm create vite@latest`)
## Workflow
1. **Discover MCP prefix** — run `list_tools`, note the Stitch prefix (e.g. `stitch:`).
2. **Fetch screen** — `[prefix]:get_screen` with project and screen IDs.
3. **Download assets** — persist HTML/screenshot under `.stitch/designs/{screen}.html` and `.png`.
4. **Read DESIGN.md** — map `colors.*`, `typography.*`, `spacing.*` to CSS variables in `src/index.css`.
5. **Generate components** — split into `src/components/`, `src/pages/`, `src/hooks/`.
6. **Wire data** — use TanStack Query for async fetches; keep presentational components pure.
## HTML → React mapping
| Pattern | Implementation |
|---------|----------------|
| Layout grid / flex | Tailwind utilities or CSS modules aligned to DESIGN.md spacing tokens |
| Cards / panels | `<section>` with tokenized border-radius and elevation fallbacks for forced-colors |
| Tables | Semantic `<table>` or TanStack Table; never div-only grids for tabular data |
| Buttons | `<button type="button">` with visible focus ring (preserve browser default unless DESIGN.md defines focus tokens) |
| Forms | `<label htmlFor>` + `<input id>`; associate errors with `aria-describedby` |
| Loading | Skeleton components; `aria-busy` on containers during fetch |
| Wallet connect | Isolate in `WalletProvider`; never embed private keys in generated code |
## DESIGN.md integration
```css
/* src/index.css — example token bridge */
:root {
--color-primary: /* from DESIGN.md colors.primary */;
--font-body: /* typography.body-md.fontFamily */;
}
```
Run the design.md linter locally before shipping UI:
```bash
npx @google/design.md lint DESIGN.md
```
## Web3 dashboard conventions
- Read-only contract calls via `useReadContract` (viem/wagmi) or ethers `Contract` + TanStack Query `queryFn`.
- Format token amounts with `formatUnits`; show network name and chain ID in settings footer.
- Surface transaction errors in plain language; link to block explorer when `txHash` exists.
- Gas-sensitive flows: batch reads, avoid redundant `eth_call` in render loops.
## File structure
```
src/
├── components/ # Presentational UI from Stitch
├── pages/ # Route-level screens
├── hooks/ # useQuery wrappers, wallet hooks
├── lib/ # ABI helpers, formatters
└── styles/ # Token CSS variables
```
## Quality checklist
- [ ] WCAG 2.2 AA: contrast from DESIGN.md component pairs passes linter
- [ ] Keyboard navigable: focus order matches visual order
- [ ] Responsive: test at 375px and 1280px widths
- [ ] No secrets in repo: RPC URLs from env (`VITE_*` prefix only for public endpoints)
- [ ] TypeScript strict: no `any` on contract ABIs
## Stitch docs note
When following links on [stitch.withgoogle.com/docs](https://stitch.withgoogle.com/docs/), use the full `https://stitch.withgoogle.com/docs/...` URL if relative navigation redirects incorrectly.Trust audit
SAFEgrade B · trust 89/100 Nothing in the source contradicts what it says it does. Grade A is reserved for packages that have also passed the behavioural sandbox.
| Layer | What it checks | Result |
|---|---|---|
| L0 | Provenance & inventory | PASS |
| L1 | Static analysis of the code | NA |
| L2 | Instruction surface (what it tells the agent) | PASS |
| L3 | Class-specific surface | PASS |
| L4 | Behavioural (sandbox) | SKIPPED |
What the source does
- Filesystem
- none-observed
- Network
- none-observed
- Shell
- none-observed
- Dependencies
- pinned
- Secrets in source
- none-found
Findings (0)
No findings outside the package's declared scope.
Gates applied: no_behavioural_pass.
32484555efabfull audit observations/trust-audit/skill/google-labs-code__react-vite-dashboard.json · Report an issue / request a re-scanAudit history
Every audit this skill has had.
| Date | Source | Verdict | Grade | Score | Change |
|---|---|---|---|---|---|
| 2026-10-22 | 32484555efab | SAFE | B | 89 | first audit |
Questions
What does the react-vite-dashboard — google-labs-code/stitch-skills skill do?
A library of Agent Skills designed to work with the Stitch MCP server. Each skill follows the Agent Skills open standard, for compatibility with coding agents such as Antigravity, Gemini CLI, Claude Code, Cursor.
Is react-vite-dashboard — google-labs-code/stitch-skills safe to install?
The audit found nothing in the source that contradicts what it says it does, and graded it B (89/100). Grade A is held back for packages that have also passed a sandboxed behavioural run, which is why a clean skill reads B.
What can react-vite-dashboard — google-labs-code/stitch-skills access on my machine?
The audit observed no filesystem, network or shell use at all in its source.
How current is this page?
The grade is for one exact copy of the source (32484555efab), read on 2026-10-22. The repository is watched, and a new audit runs when it changes — this is the first audit.
Add this badge to your README
Shows this page's audit result and links back to it. It updates when the audit does.
[](https://modelsatlas.com/skills/google-labs-code/react-vite-dashboard/)
Not your listing, or something out of date? Claim or update your listing.