Sidekick AI - Local AI Coding Assistant
100% local AI-powered coding assistant for VS Code. No internet required, no data leaves your machine.
Features
- ✅ Code Completion - Intelligent inline suggestions
- ✅ Code Explanation - Understand complex code instantly
- ✅ Code Refactoring - AI-powered improvements
- ✅ Test Generation - Automatic unit tests
- ✅ Chat Interface - Interactive AI assistant
- ✅ 100% Local - Runs entirely on your machine
- ✅ Privacy-First - No telemetry, no data collection
Requirements
- VS Code 1.74.0 or higher
- llama.cpp installed locally
- 4GB+ RAM
- A GGUF model file (instructions provided)
Setup
Install llama.cpp:
- Windows: Download from releases
- Mac/Linux:
brew install llama.cpp or build from source
Download a model:
Start the server:
llama-server.exe -m qwen-coder.gguf -c 4096 --port 8080
Configure Extension
- Install the extension
- When prompted, select your model file
- Extension will auto-detect llama.cpp
Troubleshooting
- "AI server not running": Check that llama.cpp is installed
- "Model not found": Press Ctrl+, and set the model path in settings
- Slow responses: Ensure you're using a quantized model (Q4_K_M or Q5_K_M)
Usage
- Press
Ctrl+Shift+A to open chat
- Select code and press
Alt+E to explain
- Right-click for AI options
- Type
/help in chat for commands
Privacy
This extension runs 100% locally:
- No internet connection required
- No data sent to external servers
- No telemetry or usage tracking
- All processing on your machine
License
MIT
| |
