🤖 jebat-agent

30-second setup for AI agent workspace with 8 local LLM models (Gemma 4, Qwen2.5, Hermes3, Phi-3), IDE integration, multi-agent orchestration, markdown rendering, and enterprise security.

🚀 Quick Start

# Interactive setup wizard
npx jebat-agent

# Quick setup (gateway only)
npx jebat-agent --quick

# Full setup (workspace + skills + models)
npx jebat-agent --full

# Migrate from OpenClaw/Hermes
npx jebat-agent --migrate

Live Demo: jebat.online/chat

📋 What is Jebat Agent?

Jebat Agent is your unified AI agent workspace. It combines the OpenClaw control plane and Hermes capture-first methodology into one powerful, self-hosted platform.

Features

| Feature | Description | Command | |---------|-------------|---------| | 30-Second Setup | Full workspace with skills and config | npx jebat-agent --full | | 8 Local LLMs | Deploy Gemma 4, Qwen2.5, Hermes3, Phi-3, and more | npx jebat-agent --local-model gemma4 | | IDE Integration | VS Code, Zed, Cursor, Claude Desktop, Gemini CLI | npx jebat-agent --ide vscode | | Channel Setup | Telegram, Discord, WhatsApp, Slack | npx jebat-agent --channel telegram | | Migration | Automatic OpenClaw/Hermes conversion | npx jebat-agent --migrate | | Multi-Agent | 10 core agents + 24 specialists | Via jebat-core | | 5 Orchestration Modes | Debate, Consensus, Sequential, Parallel, Hierarchical | Via chat UI | | Markdown Rendering | Tables, code blocks, headers in all responses | Automatic |

🤖 8 Local LLM Models

Run AI locally with Ollama — no API keys needed:

| Model | Size | Best For | |-------|------|----------| | Gemma 4 | 9.6GB | Best overall (default) | | Qwen2.5 14B | 9GB | Coding & reasoning | | Hermes3 | 4.7GB | Complex reasoning | | Phi-3 | 2.2GB | Fast responses (default secondary) | | Llama 3.1 8B | 4.9GB | General purpose | | Mistral | 4.4GB | Balanced performance | | CodeLlama 7B | 3.8GB | Code generation | | TinyLlama | 637MB | Lightweight testing |

🔧 Commands

Setup

# Interactive wizard
npx jebat-agent

# Quick setup (gateway only)
npx jebat-agent --quick

# Full setup (workspace + skills)
npx jebat-agent --full

# Migrate from OpenClaw/Hermes
npx jebat-agent --migrate

IDE Integration

npx jebat-agent --ide vscode    # VS Code / Cursor
npx jebat-agent --ide zed       # Zed editor
npx jebat-agent --ide claude    # Claude Desktop
npx jebat-agent --ide gemini    # Gemini CLI

Channel Setup

npx jebat-agent --channel telegram   # Telegram bot
npx jebat-agent --channel discord    # Discord bot
npx jebat-agent --channel whatsapp   # WhatsApp Business
npx jebat-agent --channel slack      # Slack app

Management

# Gateway commands
npx jebat-gateway start      # Start gateway
npx jebat-gateway stop       # Stop gateway
npx jebat-gateway restart    # Restart gateway
npx jebat-gateway status     # Check status
npx jebat-gateway logs       # View logs

# Agent commands
npx jebat-setup health       # Check health
npx jebat-setup skills       # List skills
npx jebat-setup test         # Test agent
npx jebat-setup info         # Show info

📁 Directory Structure

After running setup:

~/.jebat/
├── jebat-gateway.json    # Gateway configuration
├── .env                  # Environment variables
└── workspace/
    ├── IDENTITY.md       # Agent identity
    ├── SOUL.md           # Agent personality
    ├── TOOLS.md          # Available tools
    ├── USER.md           # User preferences
    └── skills/
        └── jebat-agent/
            └── SKILL.md  # Agent skill definition

⚙️ Configuration

Gateway config at ~/.jebat/jebat-gateway.json:

{
  "gateway": {
    "port": 18789,
    "mode": "local",
    "auth": {
      "mode": "token",
      "token": "your-token"
    }
  },
  "agents": {
    "defaults": {
      "model": {
        "primary": "ollama/gemma4:latest",
        "secondary": "ollama/phi3:latest",
        "fallbacks": [
          "ollama/qwen2.5:14b",
          "ollama/hermes3:latest"
        ]
      }
    }
  }
}

🔄 Migration

Migrating from OpenClaw/Hermes?

npx jebat-agent --migrate

This will:

• Convert ~/.openclaw → ~/.jebat
• Update all config formats
• Rename skills appropriately
• Preserve your existing setup

🌐 Ecosystem

Jebat Agent works with jebat-core for the full platform experience:

| Package | Purpose | Command | |---------|---------|---------| | jebat-agent | Setup wizard & local models | npx jebat-agent | | jebat-core | Multi-agent orchestration | npx jebat-core |

🌍 Live Demos

| Page | URL | |------|-----| | Landing | jebat.online | | Chat | jebat.online/chat | | Portal | jebat.online/portal | | Gelanggang | jebat.online/gelanggang |

💻 Requirements

• Node.js 18+
• npm/npx
• Python 3.10+ (for gateway server)
• 8GB+ RAM for local models (recommended: 16GB+)

🤝 Contributing

1. Fork the repository
2. Create your feature branch (git checkout -b feature/amazing-feature)
3. Commit your changes (git commit -m 'feat: add amazing feature')
4. Push to the branch (git push origin feature/amazing-feature)
5. Open a Pull Request

📄 License

MIT © NusaByte / JEBATCore