TripCanvas is an AI-native travel planning prototype that turns a user’s natural-language trip request into a structured, editable travel plan. Instead of behaving like a raw chatbot, it combines conversational input, itinerary generation, recommendation cards, and clear confidence/fallback states to help users move from vague travel intent to a usable trip plan. It is designed as the frontend foundation for a future booking companion rather than a one-off AI wrapper.
-
Install dependencies
pnpm install
-
Configure environment variables
Copy the example file and fill in any values your local run needs:
cp .env.example .env.local
The committed
.env.examplelists every variable the app expects. Do not commit.env.localor other files that contain secrets. For how Next.js loads env files, see the Next.js environment variables documentation. -
Start the dev server
pnpm dev
Open http://localhost:3000 in your browser.
| Command | Description |
|---|---|
pnpm dev |
Run the app in development |
pnpm build |
Production build |
pnpm start |
Start the production server |
pnpm lint |
Run ESLint |
pnpm format |
Format with Prettier |
After pnpm install, a Husky prepare script installs Git hooks; commits run lint-staged (lint + format on staged files).