Roo Code

​

Prerequisites

1. Roo Code Extension: Install Roo Code from the VS Code Marketplace
2. TrueFoundry Authentication Token: Create a Personal Access Token or Virtual Account by following the instructions in Generating Tokens
3. TrueFoundry Gateway Access: Ensure you have access to your TrueFoundry AI Gateway with the format https://your-control-plane.truefoundry.cloud/api/llm

​

Integration Guide

​

Step 1: Install Roo Code Extension

1. Open VS Code: Launch Visual Studio Code
2. Access Extensions: Click the Extensions icon in the Activity Bar (Ctrl+Shift+X)
3. Search for Roo Code: Type “Roo Code” in the search bar
4. Install Extension: Click “Install” on the Roo Code extension by RooCode

Install Roo Code from VS Code Marketplace

​

Step 2: Get TrueFoundry Gateway Configuration

1. Navigate to AI Gateway Playground: Go to your TrueFoundry AI Gateway playground
2. Access Unified Code Snippet: Use the OpenAI library code snippet
3. Copy Configuration Details: You will get the base URL and authentication details

Get Configuration from TrueFoundry Gateway

​

Step 3: Configure Roo Code with TrueFoundry Gateway

1. Open Roo Code Sidebar: Click the Roo Code icon (🦘) in the VS Code Activity Bar
2. Access Configuration: In the welcome screen or settings, look for API provider configuration
3. Select Custom Provider: Choose “Custom” or “OpenAI Compatible” from the provider dropdown
4. Configure TrueFoundry Settings:
   • Base URL: Enter your TrueFoundry Gateway URL (e.g., https://your-control-plane.truefoundry.cloud/api/llm)
   • API Key: Paste your TrueFoundry authentication token
   • Model Name: Copy the model name from our Gateway playground”

TrueFoundry Configuration in Roo Code

​

Step 4: Select AI Model

1. Choose Model: Select your preferred model from TrueFoundry’s available models
2. Get Model ID: Copy the model ID from your TrueFoundry Gateway (e.g., openai-main/gpt-4o)
3. Configure Model: Enter the model ID in Roo Code’s model selection field

Copy Model ID from TrueFoundry Gateway

• openai-main/gpt-4o: Best for complex code generation and reasoning tasks
• anthropic-main/claude-3-5-sonnet: Excellent for code understanding and following instructions
• Use the exact model ID as shown in your TrueFoundry Gateway playground

​

Step 5: Test Your Integration

1. Save Configuration: Click “Save” or “Let’s go!” to apply your settings
2. Test Connection: Open a code file and try asking Roo Code a question
3. Verify Functionality: Ensure responses are generated through TrueFoundry Gateway

Testing Roo Code with TrueFoundry

​

Advanced Configuration

​

Multiple Model Profiles

{
  "roocode.apiProfiles": [
    {
      "name": "TrueFoundry GPT-4",
      "baseUrl": "https://your-control-plane.truefoundry.cloud/api/llm",
      "apiKey": "your-truefoundry-token",
      "model": "openai-main/gpt-4o",
      "description": "High-quality code generation"
    },
    {
      "name": "TrueFoundry Fast",
      "baseUrl": "https://your-control-plane.truefoundry.cloud/api/llm", 
      "apiKey": "your-truefoundry-token",
      "model": "openai-main/gpt-3.5-turbo",
      "description": "Quick responses for simple tasks"
    }
  ]
}

​

Custom Instructions

You are an expert software developer working with enterprise-grade code.
Prioritize:
- Security best practices
- Performance optimization
- Clean, maintainable code
- Comprehensive error handling
- Detailed documentation

When suggesting code changes, always consider:
- Code review standards
- Testing requirements
- Scalability implications