nemoclaw-configure-inference - Configure NemoClaw inference providers and routing
Lists available inference providers, explains routing, validates configurations, and switches the active model without restarting the sandbox.
Tags
Updated: 2026-10-04Capabilities
Typical Inputs
Typical Outputs
What this skill does
- List inference providers
- Explain inference routing
- Switch active models
- Connect local model servers
- Validate provider configurations
- Probe tool-calling support
Inputs
- Provider selection
- Model name
- Provider API key
- Endpoint base URL
- Local model server
- Experimental mode flag
Outputs
- Provider list
- Routing explanation
- Validation result
- Active model change
- Inference route change
- Sandbox status
Requirements
- Running NemoClaw sandbox
- OpenShell CLI on PATH
- NemoClaw installed
- Running local model server or Ollama
- NEMOCLAW_EXPERIMENTAL=1 for experimental options