GolemBot: Universal Agent Deployment and Integration

GolemBot empowers developers to deploy existing coding agents like Cursor, Claude Code, and Codex across diverse communication platforms (Slack, Discord, etc.) and custom applications. It provides a unified, provider-agnostic interface, enabling teams to leverage advanced AI capabilities without the complexity of building new AI frameworks or prompt engineering, thereby democratizing access to powerful AI agents.

How It Works

The core architecture functions as a flexible gateway, abstracting away the complexities of agent execution and provider routing. Users configure their preferred coding agent and LLM provider in a single golem.yaml file. GolemBot then handles communication across various channels (IM, HTTP) and ensures a consistent StreamEvent interface, allowing seamless switching between agents and providers like OpenRouter, MiniMax, or DeepSeek with zero code changes.

Quick Start & Requirements

• Primary install: npm install -g golembot
• Core command: golembot init followed by golembot gateway or golembot run.
• Prerequisites: Node.js (version 22-slim used in Dockerfile).
• Documentation: 📖 Documentation & Guides

Highlighted Details

• Extensive Skill Ecosystem: Integrates seamlessly with ClawHub, providing access to over 13,000 community-developed AI agent skills installable via a single command.
• Universal Agent & Provider Support: Connects multiple coding agents (Cursor, Claude Code, OpenCode, Codex) with any LLM provider (OpenRouter, MiniMax, DeepSeek, etc.) through a unified configuration.
• Multi-Platform Deployment: Enables agents to operate across Slack, Telegram, Discord, Feishu, DingTalk, WeCom, WeChat, and custom Node.js applications via an embeddable API.
• Fleet Management: Offers a built-in Dashboard and Fleet Dashboard for monitoring and managing multiple running bot instances.

Maintenance & Community

The README does not provide specific details regarding notable contributors, sponsorships, or community channels like Discord/Slack.

Licensing & Compatibility

• License: MIT.
• Compatibility: Permissive MIT license allows for commercial use and integration into closed-source products.

Limitations & Caveats

Codex engine requires specific provider support for the OpenAI Responses API; providers only offering /chat/completions or /messages endpoints may not function correctly. The default codex.mode: unrestricted setting for the Codex engine bypasses sandboxing, requiring explicit configuration (safe, sandbox, approval, addDirs) for enhanced security.