An open API service indexing awesome lists of open source software.

https://github.com/semi-dlc/flowt-bb


https://github.com/semi-dlc/flowt-bb

b2b carpooling logistics ridesharing

Last synced: 1 day ago
JSON representation

Awesome Lists containing this project

README

          

# FLOWT – Sustainable Freight Ridesharing
FLOWT matches unused trucking capacity with shippers that need space. The platform combines a marketplace for offers and requests with an agent that coordinates live supply and demand, helping logistics teams collaborate, and filling empty miles.

Flowt is an experiment to determine the state of vibe-coding tools.
It is noted that while web applications are mostly created fine, the problems arising in the backend are hard to debug. Furthermore, the inherent security threats in the AI-edited code are interesting to detect.

The project is not be developed any further, and I distance myself from its sourcecode.

## TODO
- Connect to OSM to calculate route and get coordinates of places when entering data.
- Connect app to invoices and Bill of Loadings.

## Product Highlights
- B2B freight marketplace where carriers publish open capacity and shippers post their loads.
- Profile onboarding for logistics companies through Supabase.
- Rich offer/request management with validation for weights, routes, pricing, and cargo types.
- Real-time AI assistant (`supabase/functions/freight-ai-agent`) that answers route questions and recommends matches using database context.
- Developer control for tuning AI models, temperature, prompts, and token budgets.

## Tech Stack
- React 18 + Vite + TypeScript for the SPA.
- Tailwind CSS + shadcn/ui for the glassmorphism UI.
- Supabase (database, Auth, Row Level Security policies, Edge Functions).
- Deno Edge Function calling the Lovable AI gateway for freight-specific reasoning.

## Prerequisites
- Node.js 20+ and npm.
- Supabase account & CLI (to run the database locally or manage migrations).

## Getting Started
```bash
git clone
cd flowt-bb
npm install
```

1. Create a Supabase project (or run `supabase start` locally).
2. Apply database schema:
```bash
supabase db push
```
This seeds enums (`shipment_status`, `cargo_type`), tables (`profiles`, `shipment_offers`, `shipment_requests`, `bookings`), indexes, and RLS policies.
3. Copy your environment variables into `.env` (or `.env.local`):
```bash
VITE_SUPABASE_URL=
VITE_SUPABASE_PUBLISHABLE_KEY=
```
4. Start the web app:
```bash
npm run dev
```
Visit the printed URL, create an account, and complete your company profile before posting offers or requests.

## AI Agent Configuration
The AI chat panel calls the Supabase Edge Function `freight-ai-agent`. Configure its secrets so it can reach Supabase and the Lovable AI gateway:

```bash
supabase secrets set --project-ref \
LOVABLE_API_KEY= \
SUPABASE_URL= \
SUPABASE_SERVICE_ROLE_KEY=
```

Deploy the function with:
```bash
supabase functions deploy freight-ai-agent
```

The assistant automatically pulls the latest `shipment_offers`, `shipment_requests`, and `bookings` to provide contextual recommendations.

## Project Structure
```
flowt-bb/
├─ src/pages # Auth, dashboard, 404 routes
├─ src/components # Offer/request lists, dialogs, AI chat UI
├─ src/integrations # Supabase client & generated types
├─ supabase/migrations # SQL schema, RLS policies, indexes
└─ supabase/functions # Deno edge functions (freight AI agent)
```

## Useful Scripts
- `npm run dev` – start the local dev server.
- `npm run build` – production build.
- `npm run lint` – run ESLint checks.

## Next Steps
- Extend booking workflows so matched offers/requests transition through `matched → in_transit → completed`.
- Add analytics dashboards to surface fill rates and revenue impact.
- Integrate notifications (email/SMS) for new matches or AI recommendations.