Async Code Agent

Use Claude Code / CodeX CLI to perform multiple tasks in parallel with a Codex-style UI.

A code agent task management system that provides parallel execution of AI-powered coding tasks. Users can run multiple Claude Code agents simultaneously through a Codex-style web interface, with support for different agents for comparison and evaluation.

Key Features

• 🤖 Multi-Agent Support: Run Claude Code and other AI agents in parallel
• 🔄 Parallel Task Management: Execute multiple coding tasks simultaneously
• 🌐 Codex-Style Web UI: Clean interface for managing agent tasks
• 🔍 Agent Comparison: Compare outputs from different AI models
• 🐳 Containerized Execution: Secure sandboxed environment for each task
• 🔗 Git Integration: Automatic repository cloning, commits, and PR creation
• Selfhost: Deploy your own parallel code agent platform.

Architecture

• Frontend: Next.js with TypeScript and TailwindCSS
• Backend: Python Flask API with Docker orchestration
• Agents: Claude Code (Anthropic) with extensible support for other models
• Task Management: Parallel execution system based on container

Prerequisites

• Docker and Docker Compose installed
• (Optional) Supabase account for persistent data storage

Quick Start

1. 1.Clone the repository

```bash

git clone cd async-code

```

1. 1.Create the environment file

```bash

cp server/.env.example server/.env

```

Edit server/.env and set your ANTHROPIC_API_KEY. If you are using Supabase, also set SUPABASE_URL, SUPABASE_ANON_KEY and SUPABASE_SERVICE_ROLE_KEY.

1. 1.Build and start the stack

```bash

docker-compose up --build -d

```

• Frontend: http://localhost:3000
• Backend API: http://localhost:5000

Supabase Setup

1. 1.Create a new project in the Supabase dashboard.
2. 2.Open the SQL editor and run db/init_supabase.sql to create the required tables.
3. 3.Grab your project URL, anon key and service role key from Project Settings → API and place them in server/.env.

Usage

1. 1.Setup GitHub Token: Enter your GitHub token in the web interface
2. 2.Configure Repository: Specify target repository and branch
3. 3.Select Agent: Choose your preferred AI agent (Claude Code, etc.)
4. 4.Submit Tasks: Start multiple coding tasks in parallel
5. 5.Compare Results: Review and compare outputs from different agents
6. 6.Create PRs: Generate pull requests from successful tasks

Environment Variables

# server/.env
ANTHROPIC_API_KEY=your_anthropic_api_key_here

# Supabase
SUPABASE_URL=https://your-project.supabase.co
SUPABASE_ANON_KEY=your_anon_key
SUPABASE_SERVICE_ROLE_KEY=your_service_role_key

# Flask configuration
FLASK_ENV=production
FLASK_DEBUG=False

Development

# Run all services
docker-compose up