config

Files (1)

• SKILL.md

SKILL.md (first lines)

Manage the voice agent's full configuration settings.

First ensure CLI is installed:

```bash
pip install --upgrade vocal-bridge
```

## Commands

### Discover valid options

**IMPORTANT**: Before updating settings, use `vb config options` to discover valid values. This prevents errors from invalid values.

```bash
# Show all available options for current agent style
vb config options

# Show options for a specific setting (by name or label)
vb config options voice
vb config options "TTS Model"

# Show all settings in a category
vb config options audio
vb config options stt
vb config options realtime

# Show language options (for Chatty/Focused styles)
vb config options language

# Output as JSON
vb config options --json
```

### Show all settings

```bash
vb config show
```

Shows all agent settings in a readable format including:
- Core settings (style, deploy targets)
- Capabilities (background AI, hold, hangup, debug mode)
- Greeting and system prompt
- Model settings
- MCP servers
- Post-processing configuration
- Built-in integrations

For JSON output:

```bash
vb config show --json
```

### Export config sections

Export a specific config section as JSON. Output is pipe-friendly for roundtripping:

```bash
# Available sections: model-settings, client-actions, mcp-servers, api-tools, ai-agent, builtin-tools
vb config get model-settings
vb config get client-actions
vb config get ai-agent

# Save to file, edit, then re-apply
vb config get model-settings > settings.json
# edit settings.json...
vb config set --model-settings-file settings.json
```

### Partial updates with --merge

Use `--merge` to deep-merge file contents with current settings instead of replacing. Only the fields you specify are changed:

```bash
# Update only the model, keeping voice/VAD/everything else intact
echo '{"realtime": {"model": "gpt-realtime-1.5"}}' > update.json
vb config set --model-settings-file update.json --merge
```

`--merge` works with dict-based configs: `--model-settings-file`, `--builtin-tools-file`, `--ai-agent-file`. Array-based configs (client-actions, mcp-servers, api-tools) are always replaced.

### Update individual settings

```bash
vb config set [OPTIONS]
```

Available options:
- `--name NAME` - Agent name
- `--style STYLE` - Agent style (Chatty, Focused, Gemini, Ultravox, Listener)
- `--greeting TEXT` - Greeting message (use '' to clear)
- `--prompt TEXT` - System prompt (use '' to clear)
- `--deploy-targets TARGET` - Deploy targets (phone, web, both). Subscribe to Pilot to deploy on phone numbers
- `--background-enabled true|false` - Enable background AI system
- `--hold-enabled true|false` - Enable hold capability
- `--hangup-enabled true|false` - Enable hangup capability
- `--debug-mode true|false` - Enable debug event streaming
- `--post-processing-prompt TEXT` - Post-processing prompt
- `--post-processing-mcp-url URL` - Post-processing MCP server URL
- `--model-settings-file FILE` - JSON file with model settings
- `--mcp-servers-file FILE` - JSON file with MCP servers array
- `--builtin-tools-file FILE` - JSON file with built-in tools config
- `--client-actions-file FILE` - JSON file with client actions array
- `--api-tools-file FILE` - JSON file with custom HTTP API tools array
- `--max-call-duration MINS` - Max call duration in minutes (5-30)
- `--max-history-messages N` - Max messages before compaction (20-100)
- `--continuous-mode true|false` - Keep talking after short silences instead of waiting for the user each turn (tutors, narrators, guided experiences). User can still interrupt by speaking.
- `--continuous-mode-delay SECS` - Seconds to wait before auto-continuing in continuous mode (1.0-8.0, default 2.0)
- `--ai-agent-enabled true|false` - Enable AI Agent integration
- `--ai-agent-description TEXT` - Description of the developer's AI agent
- `--ai-agent-verbatim true|false` - Speak agent responses verbatim
- `--ai-agent-file FILE` - JSON file with AI Agent config
- `--outbound-enabled true|false` - Enable outbound calling (Paid subscribers only)
- `--outbound-greeting TEXT` - Greeting for outbound calls (use '' to clear)
- `--accept-outbound-tos` - Accept Outbound Calling Terms of Use (required when enabling outbound)
- `--merge` - Deep-merge file contents with current config instead of replacing (for dict-based configs)

From vocalbridgeai/vocal-bridge-claude-plugin, Apache-2.0.

More from vocalbridgeai/vocal-bridge-claude-plugin