viben gateway

Start the Gateway - the core runtime connecting channels to the agent loop.

Commands

# Start the gateway
viben gateway start

# Start with specific options
viben gateway start --host 127.0.0.1 --port 18790 --log-level info --agent main

# Start in daemon mode
viben gateway start --daemon

# Stop a running gateway
viben gateway stop
viben gateway stop --port 18790    # Stop on specific port

# Restart the gateway (stop then start)
viben gateway restart
viben gateway restart --port 18790 --log-level debug

# Check gateway status
viben gateway status

Gateway Commands

Command	Description
viben gateway start	Start the gateway, supports --host, --port, --log-level, --agent, --daemon
viben gateway stop	Stop a running gateway, supports --port to specify port
viben gateway restart	Restart the gateway (stop then start), supports all start options
viben gateway status	Check gateway running status and health

Options

Option	Description	Default
-h, --host <host>	Gateway listen address	127.0.0.1
-p, --port <port>	Gateway listen port	18790
-l, --log-level <level>	Log level (debug, info, warn, error)	info
-n, --agent <agent-id>	Specify the agent to run	main
-d, --daemon	Run in background mode	false

Architecture

┌─────────────────────────────────────────────────────────────────┐
│                           Gateway                                │
├─────────────────────────────────────────────────────────────────┤
│                                                                  │
│  ┌─────────────────────────┐    ┌────────────────────────────┐  │
│  │ Channels (Input)        │    │ Message Bus                │  │
│  │ ├── Telegram Bot        │    │ ├── Inbound Queue          │  │
│  │ ├── Discord Bot         │    │ │   (messages from         │  │
│  │ ├── WhatsApp (via       │    │ │    channels)             │  │
│  │ │   bridge)             │    │ └── Outbound Queue         │  │
│  │ ├── Feishu (WebSocket)  │    │     (responses to          │  │
│  │ └── CLI (direct input)  │    │      channels)             │  │
│  └─────────────────────────┘    └────────────────────────────┘  │
│                                                                  │
│  ┌──────────────────────────────────────────────────────────┐   │
│  │ Agent Loop                                                │   │
│  │ ├── Context Builder (system prompt + memory + skills)     │   │
│  │ ├── LLM Provider (API calls)                              │   │
│  │ ├── Tool Registry (execute tool calls)                    │   │
│  │ └── Subagent Manager (background tasks)                   │   │
│  └──────────────────────────────────────────────────────────┘   │
│                                                                  │
└─────────────────────────────────────────────────────────────────┘

Gateway Lifecycle

1. Load configuration from ~/.viben/config.yaml
2. Initialize enabled channels (Telegram, Discord, etc.)
3. Start the message bus
4. Start the agent loop
5. Process messages until shutdown

Output Examples

viben gateway start (Human-readable):

Gateway starting...
  Agent: main
  Model: claude-sonnet-4-20250514
  Channels:
    ✓ telegram    connected   @my_bot
    ✓ discord     connected   MyBot#1234
    ○ whatsapp    disabled
    ○ feishu      disabled

Gateway running. Press Ctrl+C to stop.

viben gateway start --json:

{
  "success": true,
  "data": {
    "status": "running",
    "agent": "main",
    "model": "claude-sonnet-4-20250514",
    "channels": [
      {"name": "telegram", "status": "connected", "identifier": "@my_bot"},
      {"name": "discord", "status": "connected", "identifier": "MyBot#1234"}
    ],
    "pid": 12345
  }
}

viben gateway status (Human-readable):

Gateway Status: Running

  Host: 127.0.0.1
  Port: 18790
  PID:  12345
  Agent: main
  Uptime: 2h 15m

Channels:
  telegram    connected   @my_bot        2h ago
  discord     connected   MyBot#1234     2h ago

Health:
  ✓ Message bus: healthy
  ✓ Agent loop: healthy
  ✓ Memory usage: 128MB

API Endpoints

The gateway provides the following HTTP APIs after startup:

Endpoint	Description
GET /health	Health check
GET /api/agent	Agent management
GET /api/sessions	Session management
POST /api/chat	Chat API
GET /api/providers	Provider configuration
GET /api/models	Model configuration

Default address: http://127.0.0.1:18790

Related Commands

• viben channel - Channel management
• viben cron - Scheduled task management
• viben agent - Agent management