VSCode

Configure VSCode to use CodeFreedom's proxy for AI features.

Quick Start

# Generate VSCode config
cf setup config vscode
# or
cf s c vscode

This generates ~/.codefreedom/vscode/settings.json with proxy settings.

How It Works

VSCode → localhost:4000 → LiteLLM Proxy → AI Model

VSCode talks to your local proxy. The proxy routes requests to whichever model you configured.

Configuration

The generated config is at ~/.codefreedom/vscode/settings.json:

{
  "github.copilot.chat.codeGeneration.models": {
    "custom": {
      "displayName": "Qwen3.6-27B",
      "requestUrl": "http://localhost:4000/v1/chat/completions",
      "model": "custom:Qwen3.6-27B"
    }
  }
}

Apply to VSCode

Copy the generated config to your VSCode settings:

cp ~/.codefreedom/vscode/settings.json ~/.config/Code/User/settings.json

Or use VSCode's settings UI to import the config.

Troubleshooting

VSCode Can't Connect

# Check proxy is running
cf run proxy status
# or
cf r px status

# Test connectivity
curl http://localhost:4000/v1/models