Skip to content

AbduTello/CodeCrew---HD4

Repository files navigation

CodeCrew - HackDearborn 4

tag:innovationlab tag:hackathon

Multi-agent system where specialized AI agents collaborate to automate the full software development process: writing code, running tests, and maintaining project context. Built with Fetch.ai's uAgents framework and powered by ASI:One.

** Built for Hack Dearborn 2025 - Fetch.ai Track**


Project Overview

CodeCrew is an autonomous software development system powered by 6 specialized AI agents that communicate via Fetch.ai's chat protocol and use ASI:One for intelligent decision-making. Simply describe your project, and watch as our agents collaborate to build it in real-time.

Key Features

  • Multi-Agent Collaboration: 6 specialized agents work together seamlessly
  • Real-time Updates: Live WebSocket connection shows agent activity
  • Live Preview: Instantly preview generated web applications
  • Full Stack Generation: Frontend, backend, tests, and documentation
  • Powered by Fetch.ai: All agents deployed on Agentverse with chat protocol

The Agent Team

  1. Orchestrator Agent - Coordinates tasks and maintains context
  2. Developer Agent - Writes code using ASI:One reasoning
  3. Tester Agent - Reviews code and writes tests
  4. Documentation Agent - Auto-generates documentation
  5. UI Agent - Specializes in frontend/React components
  6. API Agent - Specializes in backend/API development

Quick Start

Prerequisites

  • Python 3.9+
  • Node.js 18+
  • npm or yarn
  • Git

1. Clone the Repository

git clone https://github.com/yourusername/codecrew.git
cd codecrew

2. Backend setup

# Create virtual environment
python -m venv venv
source venv/bin/activate  # On Windows: venv\Scripts\activate

# Install dependencies
pip install -r requirements.txt

# Configure environment variables
cp .env.example .env
# Edit .env with your API keys (see Configuration section)

3. Frontend setup

cd codecrew-ui

# Install dependencies
npm install

# Configure environment
cp .env.example .env
# Edit .env if needed (default values should work locally)

cd ..

4. Run the Application

# Terminal 1 - Backend:
source venv/bin/activate
cd backend
python main.py

# Terminal 2 - Frontend:

cd codecrew-ui
npm run dev

Visit: http://localhost:5173


Backend Environment Variables (.env in root)

# Fetch.ai Credentials
AGENTVERSE_API_KEY=your_agentverse_api_key_here

# ASI:One API (get from hackpack)
ASI_ONE_API_KEY=your_asi_one_api_key_here

# LLM API Keys (choose one)
ANTHROPIC_API_KEY=your_anthropic_key_here
OPENAI_API_KEY=your_openai_key_here

# Backend Configuration
PORT=8000

Frontend Environment Variables (codecrew-ui/.env)

VITE_API_URL=http://localhost:8000
VITE_WS_URL=ws://localhost:8000/ws

Getting API Keys

  1. Agentverse: Sign up at agentverse.ai
  2. ASI:One: Access from Hack Dearborn hackpack
  3. Anthropic/OpenAI: Get from respective provider websites

For New Team Members

Clone the repo and create a branch:

  1. git clone (LINK)
  2. cd codecrew
  3. git checkout -b feature/your-name
  4. Get API keys from team lead (don’t commit these!)
  5. Follow Quick Start guide above

Technology Stack

Backend

  1. uAgents - Fetch.ai's agent framework

  2. FastAPI - Modern Python web framework

  3. WebSockets - Real-time communication

  4. ASI:One - AI reasoning engine

Frontend

  1. React - UI library

  2. Vite - Fast build tool

  3. Tailwind CSS - Utility-first CSS

  4. Lucide React - Icon library

  5. React Syntax Highlighter - Code display

AI & Agents

  1. Agentverse - Agent deployment platform

  2. Chat Protocol - Inter-agent communication

  3. ASI:One - LLM reasoning engine

About

Multi-agent system for automated software development. Specialized AI agents collaborate to write code, run tests, and manage projects autonomously using Fetch.ai.

Resources

License

Contributing

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors