async-code  by ObservedObserver

Parallel AI code agent system

Created 3 months ago
405 stars

Top 71.8% on SourcePulse

GitHubView on GitHub
1 Expert Loves This Project
Project Summary

This project provides a self-hostable, parallel AI code agent system with a Codex-style web UI, enabling users to run multiple AI coding tasks concurrently for comparison and evaluation. It targets developers and researchers seeking to manage and compare outputs from different AI models like Claude Code, streamlining workflows with features like Git integration and containerized execution.

How It Works

The system employs a Python Flask backend with Docker orchestration for parallel task management and containerized execution of AI agents. A Next.js frontend provides a Codex-style UI for task submission and output comparison. Tasks are executed in parallel, with support for multiple AI models, allowing users to evaluate different agents' performance on the same coding tasks within isolated, sandboxed environments.

Quick Start & Requirements

  • Install via git clone and docker-compose up --build -d.
  • Prerequisites: Docker, Docker Compose. Optional: Supabase account for persistent data.
  • Requires ANTHROPIC_API_KEY and optionally Supabase credentials set in server/.env.
  • Setup involves cloning, creating .env, configuring API keys, and running docker-compose.
  • Official quick-start and usage details are available in the repository.

Highlighted Details

  • Multi-agent support for running Claude Code and other AI agents in parallel.
  • Parallel task management for simultaneous execution of coding tasks.
  • Codex-style web UI for intuitive task management and output comparison.
  • Containerized execution provides a secure, sandboxed environment for each task.
  • Integrated Git support for automatic cloning, commits, and Pull Request creation.

Maintenance & Community

No specific contributors, sponsorships, or community links (Discord/Slack) are mentioned in the README.

Licensing & Compatibility

The project's license is detailed in the LICENSE file. No specific license type is stated in the README, and compatibility for commercial or closed-source use is not explicitly mentioned.

Limitations & Caveats

The project requires setting up API keys for AI models and potentially Supabase for data persistence. While it supports multiple agents, the README primarily highlights Claude Code, with extensibility for others noted.

Health Check
Last Commit

1 month ago

Responsiveness

Inactive

Pull Requests (30d)
0
Issues (30d)
1
Star History
62 stars in the last 30 days

Explore Similar Projects

Feedback? Help us improve.