Hex Quick Responder

HexProperty's Automated Dialog Response System with AI Integration

Features

• Automatically respond to common VS Code dialogs
• AI-powered responses using multiple LLM providers
• Configurable provider selection and model choices
• Cost tracking for AI usage
• Custom provider support through JSON configuration
• Secure API key management through environment variables

Setup

1. Install the extension
2. Copy .env.example to .env and add your API keys:

   OPENROUTER_API_KEY=your_openrouter_api_key_here
   TOGETHER_API_KEY=your_together_api_key_here
   ANTHROPIC_API_KEY=your_anthropic_api_key_here
   

Configuration

Extension Settings

• hexQuickResponder.autoRespond: Enable/disable automatic dialog responses
• hexQuickResponder.useAi: Enable/disable AI processing for unknown dialogs
• hexQuickResponder.selectedProvider: Selected LLM provider for AI processing
• hexQuickResponder.selectedModel: Selected model for the current provider
• hexQuickResponder.responses: Mapping of questions to their automatic responses
• hexQuickResponder.customProviders: List of custom LLM provider configurations

Default Providers

1. OpenRouter.ai

   • Models: Qwen 32B, Claude 2
   • Features: Wide model selection, competitive pricing

2. Together.ai

   • Models: Llama 2 70B
   • Features: Open source models, cost-effective

3. Anthropic Direct

   • Models: Claude 2, Claude Instant
   • Features: High performance, extensive context window

Adding Custom Providers

1. Command Palette > "Hex: Add Custom LLM Provider"
2. Enter provider configuration in JSON format:

   {
     "id": "custom",
     "name": "Custom Provider",
     "baseUrl": "https://api.custom-provider.com",
     "headerTemplate": {
       "Content-Type": "application/json"
     },
     "models": [
       {
         "id": "model-1",
         "name": "Model One",
         "contextLength": 4096,
         "costPer1kTokens": 0.001,
         "description": "Description of the model"
       }
     ],
     "defaultModel": "model-1"
   }

3. Add corresponding API key to .env file:

   CUSTOM_API_KEY=your_api_key_here
   

Commands

• Hex: Quick Respond to Dialog (Ctrl+Alt+H): Manually trigger dialog response
• Hex: Add Quick Response Mapping: Add new question/response mapping
• Hex: Add Custom LLM Provider: Add new provider configuration

Three Iterations Methodology

The AI responses follow a structured approach:

1. Understand the core problem
2. Break it down into actionable steps
3. Execute with clear direction

This ensures responses are focused on getting things done efficiently.

Security

• API keys are stored securely in the .env file
• Environment variables are used to prevent key exposure
• Custom provider configurations are stored in VS Code settings

Cost Management

• Cost per request is displayed after each AI response
• Token usage tracking helps monitor API consumption
• Model selection allows choosing cost-effective options