Cognito

Modern AI-powered agentic e-commerce platform, designed to autonomously handle complex commerce operations.

Open Source Software - Licensed under the Functional Source License (FSL). Free to use and modify.

Demo app link: TBA

Design Vision

Below are the reference designs that showcase how the platform should look and feel:

Home Page

Product Listing

Product Page

For more details about the template system and customization, see template/README.md.

Project Overview

Cognito is an innovative agentic e-commerce platform that leverages artificial intelligence and autonomous agents to enhance the shopping experience. Unlike traditional e-commerce systems, Cognito uses AI agents to handle complex workflows, customer interactions, and business processes autonomously.

AI Agents Architecture

The platform uses LangGraph to implement a multi-agent system where specialized AI agents autonomously handle different aspects of e-commerce operations. For detailed architecture documentation, see docs/AGENTS.md.

AI Ideas & Planned Features

The platform will leverage multiple AI models to create an intelligent, safe, and autonomous e-commerce experience:

Conversational Ecommerce Experience

• AI Chat Interface - Natural language product discovery and purchase through conversational AI
• Personalized Recommendations - Context-aware product suggestions based on conversation
• Smart Shopping Assistant - Autonomous agent to guide customers through their shopping journey

Content Moderation

• Qwen Guard - Vulgar language detection for product submissions in demo store
• Qwen Guard - Comment filtering and moderation for inappropriate content

Product Management

• Qwen 8 VL - Automatic image analysis and product description generation
• Neural Network Classifier - Intelligent filter population and product categorization

Inventory Personal Assistance

• AI Agent - Supplier analysis and cost optimization to find cheaper suppliers
• Competitive Analysis - Automated competitor monitoring and pricing insights
• Inventory Advisory - Intelligent recommendations for stock management and reordering strategies

Main Components

1. API Backend (Agentic)

   • Built with LangGraph - framework for building stateful, multi-agent AI applications
   • Autonomous AI agents handle complex e-commerce workflows and business processes
   • MongoDB for application data storage
   • Weaviate for vector embeddings and semantic search
   • Multi-agent orchestration for order processing, inventory management, and customer service

2. AI Chat

   • Natural language product search
   • Intelligent recommendations
   • Direct purchase capability through chat
   • Mobile-first responsive design with desktop support

3. CMS

   • Admin panel for store configuration
   • Product management
   • Order and inventory management
   • Personalization and settings
   • Mobile-first responsive design with desktop support

Technologies

• Frontend: Next.js + TypeScript
• Backend: LangGraph + MongoDB
• AI: LangGraph for conversational commerce
• Database:
  • MongoDB - Primary database for application data
  • Weaviate - Vector database for AI-powered search and recommendations
• i18n: next-intl (English and Polish support)

Status

Project in initialization phase.

MVP Progress

Feature	Status
Login	✅
Simple Product Search	✅
Checkout	❌
Payments	❌
Add New Products	✅
Store Configuration	❌
Browse Orders	❌
Browse Users	❌
Other	❌

Running the Project

Quick Start (Easiest)

Using Make (Recommended):

# Start everything (Docker + Next.js) in one command
make dev-full

Using npm:

npm run dev:full

This will start:

• Docker services (MongoDB, Weaviate, vLLM)
• Next.js development server

Manual Start

Step 1: Start Infrastructure

# Using Make
make dev-infra

# OR using npm
npm run docker:up

# OR using docker-compose directly
docker-compose up -d

Step 2: Start Development Server

# Using Make
make dev

# OR using npm
npm run dev

Available Commands

Make commands (easier to remember):

make help         # Show all available commands
make dev-full     # Start everything
make dev          # Start only Next.js
make dev-infra    # Start only Docker services
make stop         # Stop Docker services
make logs         # View Docker logs
make restart      # Restart Docker services
make clean        # Remove Docker volumes
make test         # Run tests
make build        # Build for production

npm scripts:

npm run dev           # Next.js dev server only
npm run dev:full      # Docker + Next.js
npm run docker:up     # Start Docker services
npm run docker:down   # Stop Docker services
npm run docker:logs   # View logs
npm run test          # Run tests
npm run build         # Build for production

Available Services

• Application: http://localhost:2137
  • Shop (AI Chat): http://localhost:2137/en/shop/chat
  • CMS (Admin): http://localhost:2137/en/cms
• MongoDB: localhost:2138
• Weaviate: localhost:2139 (HTTP), localhost:2140 (gRPC)
• vLLM Qwen3-VL: localhost:2141
• vLLM Qwen3: localhost:2142

Production Deployment

# Build and start all containers (app + infrastructure)
docker-compose -f docker-compose.prod.yml up -d

Configuration

Copy .env.example to .env and adjust environment variables:

cp .env.example .env

cognito