A browser-based data analysis and plotting app: tables, statistical analyses, and graphs (bar, scatter, line, dose-response, etc.) with an optional Chat panel driven by an LLM (Groq, Anthropic, or any OpenAI-compatible API).
Stack: React 19, TypeScript, Vite 7, Zustand, Plotly.js, Vercel AI SDK.
npm install
cp .env.example .env.local # optional: set one LLM key for Chat
npm run devOpen the URL shown (e.g. http://localhost:5173/). Use the sidebar to add tables, analyses, and graphs; use the right-hand Chat panel (with an API key set) for natural-language commands.
Docs follow the Diataxis framework and live in docs/:
| Purpose | Where to look |
|---|---|
| Learn / run | docs/tutorials/getting-started.md |
| Do a task | docs/how-to-guides/ |
| Look up | docs/reference/ (layout, env, architecture) |
| Understand | docs/explanation/ (design decisions) |
Coding agents and contributors: read AGENTS.md for conventions and where to change things.
| Command | Description |
|---|---|
npm run dev |
Start dev server |
npm run build |
Type-check + production build |
npm test |
Run Vitest tests |
npm run lint |
Run ESLint |
See LICENSE if present.