Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
Version History
- Updated mode-switching mechanism: now uses "gateway config.patch" to persistently change the primary model in openclaw.json (triggers a brief gateway restart). - Improved status reporting: mode status now uses session_status and displays a formatted summary of all modes, including active/inactive indicators. - Clarified instructions and examples for setup, switching, and status commands. - Adjusted notes to emphasize persistent config changes and brief restart after switching modes.
What This Skill Does
Lets you switch between four AI model tiers (eco, balanced, smart, max) using short text commands. Each tier maps to a specific model ID stored in a local config file. The active model updates instantly without restarting.
Instead of manually editing config files each time you need a different model, a single command handles the read-write cycle and confirms the switch.
When to use it
- Routing simple questions to a cheaper model to cut API costs
- Switching to a high-power model for complex multi-step reasoning
- Checking which model tier is active before starting a session
- Configuring which model each tier uses on first setup
- Dropping back to a lighter model after finishing a demanding task
Example Workflow
Here's how your AI assistant might use this skill in practice.
User says: eco mode
- 1Reads ~/.openclaw/workspace/switch-modes.json to find the model mapped to 'eco'
- 2Reads ~/.openclaw/openclaw.json to load the current config
- 3Updates the model field with the eco mode model ID
- 4Writes the updated config back to ~/.openclaw/openclaw.json
- 5Confirms the change to the user
✅ ECO mode activated Now using: anthropic/claude-3.5-haiku