claw-empire  by GreenSheep01201

AI agent office simulator for autonomous company management

Created 1 month ago
1,030 stars

Top 36.2% on SourcePulse

GitHubView on GitHub
Project Summary

Claw-Empire provides a local-first AI agent office simulator, enabling users to orchestrate a diverse array of CLI, OAuth, and API-connected AI agents as a virtual company. It targets developers and power users seeking a unified, private, and visually intuitive platform to manage AI collaboration, offering a unique "CEO desk" perspective for overseeing autonomous agent operations. The primary benefit is centralized control and transparent visualization of complex AI workflows without relying on external cloud services.

How It Works

Claw-Empire transforms AI coding assistants into virtual employees within a simulated company, visualized through a pixel-art office interface. Agents collaborate across departments, attend meetings, and deliver tasks, all managed locally via an SQLite database. The system supports agents connected through CLI tools, OAuth providers (like GitHub/Google), and direct API keys to external LLMs. Core to its design is agent isolation within git worktrees for safe parallel development and a local-first architecture ensuring data privacy.

Quick Start & Requirements

  • Primary Install: Clone the repository and run the platform-specific install script:
    • macOS/Linux: git clone https://github.com/GreenSheep01201/claw-empire.git && cd claw-empire && bash install.sh
    • Windows (PowerShell): git clone https://github.com/GreenSheep01201/claw-empire.git; cd claw-empire; powershell -ExecutionPolicy Bypass -File .\install.ps1
  • Run Command: pnpm dev:local
  • Prerequisites: Node.js (>= 22), pnpm (latest), Git.
  • Docs: AI Install Guide, Release Notes, API Docs.

Highlighted Details

  • Local-first AI agent orchestration with a pixel-art office simulator.
  • Supports a wide range of AI providers via CLI, OAuth, and direct API keys.
  • Multiple workflow "pack profiles" tailor agent collaboration for different tasks (e.g., development, reporting, creative writing).
  • Integrated messenger support (Telegram, Discord, Slack, etc.) for CEO directives.
  • Secure, local storage of OAuth and messenger tokens using AES-256-GCM encryption.
  • Agents operate in isolated git worktrees, merging only upon CEO approval.

Maintenance & Community

The project shows signs of active development with recent release notes (v2.0.1). Contributions are welcomed via pull requests to the dev branch, following guidelines in CONTRIBUTING.md. Specific community channels like Discord or Slack are not explicitly mentioned in the README.

Licensing & Compatibility

The project is licensed under the Apache 2.0 license, permitting free personal and commercial use.

Limitations & Caveats

While designed for robustness, the auto-update feature details troubleshooting steps for potential git_pull_failed errors or diverged local branches, indicating that manual intervention may sometimes be required to maintain the repository. The AUTO_UPDATE_RESTART_COMMAND has strict security limitations, disallowing shell metacharacters and direct shell launchers.

Health Check
Last Commit

3 weeks ago

Responsiveness

Inactive

Pull Requests (30d)
9
Issues (30d)
4
Star History
546 stars in the last 30 days

Explore Similar Projects

Starred by Andrew Ng Andrew Ng(Founder of DeepLearning.AI; Cofounder of Coursera; Professor at Stanford), Jack Lukic Jack Lukic(Author of Semantic UI), and
5 more.

ag2 by ag2ai

0.7%
4k
AgentOS for building AI agents and facilitating multi-agent cooperation
Created 1 year ago
Updated 11 hours ago
Feedback? Help us improve.