Skip to main content
The /new command creates a new deployment configuration through an interactive multi-step wizard. It collects all necessary information for deploying an OpenClaw instance.

​
Usage

/new

​
Configuration Steps

The wizard guides you through these steps:

​
1. Template Selection

Choose to start from a template or create a blank configuration: 
Step 1: Choose Template
Start from a template or blank configuration (use arrow keys and Enter):

┌────────────────────────────────────────┐
│ > Blank configuration                  │
│   Anthropic Claude (Telegram) [built-in] │
│   OpenAI GPT-4 (Telegram) [built-in]  │
└────────────────────────────────────────┘

​
2. Deployment Name

Enter a unique name (lowercase, alphanumeric, hyphens allowed): 
Step 2: Deployment Name
Enter a unique name for this deployment:

Name: my-openclaw-server

​
3. Cloud Provider

Select your cloud provider: 
Step 3: Cloud Provider
Select where to deploy (use arrow keys and Enter):

┌────────────────────────────────────────┐
│ > Hetzner Cloud - Recommended - US East│
│   DigitalOcean - NYC1 - New York       │
└────────────────────────────────────────┘

​
4. API Key

Choose a saved API key or enter a new one: 
Step 4: Hetzner API Key
Choose a saved key or enter a new one:

┌────────────────────────────────────────┐
│ > Enter a new API key                  │
│   prod-key (hetzner_...)               │
└────────────────────────────────────────┘
If entering a new key:
  • The key is validated against the provider’s API
  • You’ll be prompted to save it for future use
  • Saved keys are stored securely in ~/.clawcontrol/secrets/

​
5. Droplet Size (DigitalOcean Only)

Select your droplet size: 
Step 5: Droplet Size
Select your DigitalOcean droplet size:

NOTE: Minimum size for OpenClaw is 2GB RAM

┌────────────────────────────────────────┐
│ > s-1vcpu-2gb - 1 vCPU, 2GB RAM - $12/mo │
│   s-2vcpu-2gb - 2 vCPU, 2GB RAM - $18/mo │
│   s-2vcpu-4gb - 2 vCPU, 4GB RAM - $24/mo │
└────────────────────────────────────────┘

​
6. AI Provider

Select your AI model provider: 
Step 6: AI Provider
Select your AI model provider:

┌────────────────────────────────────────┐
│ > Anthropic - Claude models            │
│   OpenAI - GPT models                  │
│   Google - Gemini models               │
│   Groq - Fast inference                │
│   OpenRouter - Multi-provider gateway  │
└────────────────────────────────────────┘

​
7. AI API Key

Provide your AI provider API key (can be saved for reuse).

​
8. Model

Enter the model identifier: 
Step 8: Default Model
Enter the model identifier for Anthropic.

e.g. claude-sonnet-4-20250514, claude-3-5-haiku-20241022

Model: claude-sonnet-4-20250514

​
9. Telegram Bot Token

Provide your Telegram bot token (from @BotFather).

​
10. Telegram User Access

Specify which Telegram user IDs can access the bot: 
Step 10: Telegram User ID
Enter your Telegram user ID for access control.

User ID: 123456789

​
11. Tailscale Configuration

Choose whether to configure Tailscale VPN: 
Step 11: Tailscale
Select Tailscale configuration:

┌────────────────────────────────────────┐
│ > Skip                                 │
│   Configure Tailscale                  │
└────────────────────────────────────────┘

​
12. Confirmation

Review and confirm your configuration: 
Deployment Configuration Summary

Name:           my-openclaw-server
Provider:       hetzner
Server:         CPX11 (2 vCPU, 2GB RAM)
Location:       Ashburn, VA (US East)
AI Provider:    anthropic
Model:          claude-sonnet-4-20250514
Channel:        telegram
Tailscale:      Enabled

Press Y to confirm, N to go back

​
Edit Mode

You can edit existing deployments from the /list command. Edit mode skips some steps (name, provider) and pre-fills existing values.

​
Fork Mode

Fork an existing deployment to create a copy with a new name. Use this from /list or /templates.

​
Key Features

  • API Key Validation: All provider and AI keys are validated in real-time
  • Secure Storage: API keys and tokens are stored securely
  • Key Reuse: Save keys once and reuse across multiple deployments
  • Template Support: Start from built-in or custom templates
  • Flexible Navigation: Use Esc to go back, arrow keys to navigate

​
Tips

Use templates (via /templates) to create reusable presets for common configurations. This saves time when creating multiple similar deployments.
The minimum recommended RAM for OpenClaw is 2GB. Smaller instances may experience out-of-memory errors.

​
Example Terminal Output

clawcontrol v0.2.3

> /new

Step 1: Choose Template
┌────────────────────────────────────────┐
│ > Blank configuration                  │
│   Anthropic Claude (Telegram) [built-in] │
└────────────────────────────────────────┘

[Enter]

Step 2: Deployment Name
Name: production-claude

[Enter]

Validating API key...
✓ API key validated successfully

Step 4: Hetzner API Key
Would you like to save this API key for future deployments?
Press Y to save, N to skip

[Y]

Key Name: prod-hetzner
✓ Key saved

...

Deployment created successfully!
Run /deploy to deploy to the cloud.

​
Next Steps

After creating a deployment configuration:
  1. Review it with /list
  2. Deploy it with /deploy
  3. Monitor with /status

Deploy Command

Deploy your configuration to the cloud

Build docs developers (and LLMs) love

Get started for freeTalk to us