How to Integrate Kilocode Models in Openclaw

Table of contents: [Show]

Quick Start

Kilocode is an AI coding assistant platform that provides access to powerful language models. This guide shows you how to integrate Kilocode models into your Openclaw setup in under 5 minutes.

Get Your API Key

Go to app.kilo.ai
Sign in or create an account
Navigate to API Keys and generate a new key

Configure Openclaw

To make Kilocode your primary agent model, edit your ~/.Openclaw/Openclaw.json configuration file:

{
  "env": { "KILOCODE_API_KEY": "<your-kilocode-api-key>" },
  "agents": {
    "defaults": {
      "model": { "primary": "kilocode/kilo/auto" }
    }
  }
}

After saving the configuration, restart your Openclaw gateway to apply the changes:

openclaw gateway restart

For more deployment options, check out our guide on Flyio Deployment. If you are new to Openclaw, you might also want to read about Elevated Mode for advanced permissions.

Testing Your Setup

Once configured, test the integration by sending a prompt:

openclaw message --content "Explain how Kilocode models work"

The response should come from the Kilocode model, confirming your setup is working correctly.

Troubleshooting

Authentication errors: Verify your API key is correctly copied and the Openclaw.json syntax is valid JSON
Model not found: Ensure you are using the correct model ID: kilocode/kilo/auto
Gateway not responding: Check that the gateway restarted successfully with openclaw gateway status

Best Practices

Store your API key in environment variables for production deployments
Use kilocode/kilo/auto for automatic model selection based on task complexity
Monitor API usage in your Kilocode dashboard to track costs