The onboard command sets up Weaver for first-time use by creating the configuration file and workspace directory with default templates.
Usage
What It Does
- Creates configuration at
~/.weaver/config.json
- Creates workspace directory with embedded templates
- Prompts for overwrite if config already exists
- Displays next steps for API key setup
Interactive Example
$ weaver onboard
🕸️ weaver is ready!
Next steps:
1. Add your API key to /home/user/.weaver/config.json
Get one at: https://openrouter.ai/keys
2. Chat: weaver agent -m "Hello!"
Generated Files
Configuration File
Location: ~/.weaver/config.json
Contains:
- Provider settings (OpenRouter, Anthropic, OpenAI, etc.)
- Agent defaults (model, temperature, max tokens)
- Gateway configuration (host, port)
- Channel settings (Telegram, Discord, Slack)
- Workspace path
Workspace Directory
Default location: ~/.weaver/workspace/
Structure:
workspace/
├── skills/ # Custom agent skills
├── cron/ # Scheduled task definitions
│ └── jobs.json
└── state/ # Runtime state storage
Exit Codes
| Code | Reason |
|---|
| 0 | Success - config and workspace created |
| 1 | Failed to save config or create workspace |
Error Conditions
Error saving config: permission denied
~/.weaver/
Error copying workspace templates: ...
Next Steps
After onboarding:
Add API Key
Edit ~/.weaver/config.json and add your OpenRouter API key:"openrouter": {
"api_key": "sk-or-..."
}
Test Agent
Verify setup with a test message: Start Gateway (Optional)
Launch the gateway server for channel integrations: Configuration Override
To re-run onboarding and overwrite existing config:
$ weaver onboard
Config already exists at /home/user/.weaver/config.json
Overwrite? (y/n): y
🕸️ weaver is ready!
Overwriting will reset all custom configuration. Back up your config first if you have custom settings.