How karanxrathod/annadata-ai works
Project type and purpose
• This is a full-stack web application starter built with Next.js (app router) and designed to deploy on Firebase Hosting (apphosting.yaml).
• The primary features are three AI-driven “flows” (crop diagnosis, government schemes navigation, real-time market analysis), surfaced through React UI components and powered by a local GenKit/flows library.
File structure
• Top-level:
• next.config.ts, tsconfig.json, tailwind.config.ts, postcss.config.mjs – client/server build and styling configuration
• apphosting.yaml – Firebase Hosting configuration
• package.json / package-lock.json – dependencies and scripts
• src/app – Next.js “app router” pages and global layout (layout.tsx, page.tsx, globals.css)
• src/ai – server-side AI logic and orchestration
• genkit.ts – shared AI helper library
• flows/*.ts – individual AI workflows
• src/components – React UI library
• features – feature-specific pages (crop-diagnosis, govt-schemes, market-analysis)
• layout – global header/footer
• ui – atom- and molecule-level building blocks (buttons, accordions, dialogs, etc.)
• src/hooks & src/lib – custom React hooks and utility functions
• docs – markdown describing system blueprints
README
• The README simply states “NextJS starter in Firebase Studio” and points to src/app/page.tsx as the entry point.
Detail & coverage
• Don’t just show “frontend” and “backend.” Break out:
• UI Library vs. Feature Components
• Next.js Routing vs. Server Components vs. API Routes
• AI GenKit core vs. individual flow modules
• External AI provider vs. Firebase runtime
• Data flow: user → UI → server → AI → server → UI → user.