Involvex Smart Autocomplete

Involvex Smart Autocomplete is a cutting-edge VS Code extension that delivers intelligent, context-aware tab completions by dynamically integrating multiple AI models and providers.

Whether you rely on powerful cloud models like OpenAI and Anthropic or prefer privacy-focused local LLMs (via Ollama, LM Studio), this extension adapts to your workflow, learns your patterns, and helps you code faster.

🚀 Features

🧠 Adaptive Intelligence

• Multi-Context Awareness: analyzes your active file, cursor position, syntax, and project structure.
• Project-Wide Analysis: Understands cross-file dependencies and repository conventions.
• Git Awareness: Prioritizes completions relevant to recent changes and active branches.

🔌 Provider Agnostic

• Seamless Switching: Toggle between OpenAI, Anthropic, Gemini, Mistral, and Local LLMs instantly.
• Fallback Mechanisms: Automatically switches providers if one hits rate limits or goes down.
• Local-First Support: First-class support for llama.cpp, Ollama, and LM Studio.

⚡ Performance & Privacy

• Smart Caching: Caches frequent requests to minimize latency and cost.
• Privacy Controls: Opt-in telemetry. Your code stays yours.
• Debounced Triggers: Auto-activates only in high-confidence scenarios to avoid distraction.

🛠️ Tech Stack using Bun

This project is built using Bun for fast dependency management and scripting.

• Funtion: VS Code Extension API & LSP
• Inference: WebAssembly & HTTP Streaming
• Language: TypeScript

📦 Installation

1. Install the extension from the VS Code Marketplace.
2. Open your project.
3. Configure your API keys in the extension settings.
4. Start coding!

📖 Documentation

• Setup Guide - Installation and initial configuration.
• Provider Configuration - How to set up OpenAI, Anthropic, and Local models.
• Prompt Engineering - Customize system prompts and templates.

⌨️ Commands

• Smart Autocomplete: Toggle Status - Enable/Disable the extension.
• Smart Autocomplete: Switch Provider - Quickly change active AI provider.
• Smart Autocomplete: Clear Cache - Reset learned patterns and cache.

🤝 Contributing

We welcome contributions! Please see our Contributing Guide for details.

1. Clone the repository.
2. Run bun install.
3. Run bun run compile.
4. Press F5 to start debugging.

📄 License

MIT