🚀 Upscale - A VS Code Extension

Upscale: Auto-Refactor & Upskill Upscale helps you polish and modernize your work as you go, ensuring your code is as sharp as your ideas.

🖥️ Upscale Preview

Simply highlight the code block you want to "Upskill". Right-click and select Upscale from the navigation menu:

🛠 What is Upscale?

Upscale is a VS Code extension designed to bridge the gap between "functional" code and "architectural" excellence. Unlike generic AI tools, Upscale analyzes your code selection and applies specific 2026 industry standards to refactor your logic into cleaner, faster, and more maintainable snippets.

Key Features

• Smart Context Analysis: Captures highlighted code and language metadata to provide the AI with precise situational awareness.
• Architectural Personas: Every refactor is performed by a specialized Senior Architect persona, focusing on performance, scalability, and modern syntax.
• Dynamic Knowledge Base: Uses a custom docs.ts provider to inject framework-specific standards (like Hono) directly into the AI's logic.

🏗 How It Works

1. Selection: Highlight a messy code block or just click a line.
2. Context Extraction: The extension gathers the language ID and current text selection for the AI.
3. Refactor Engine: Our llm.ts core communicates with the Gemini API to process the architectural "Upscale".
4. Instant Delivery: Optimized code is returned instantly as a VS Code notification.

🚀 Getting Started

Prerequisites

• VS Code version ^1.110.0.
• A Gemini API Key (get one for free at Google AI Studio).

Installation & Setup

1. Clone this repository into your local environment.
2. Run npm install to set up dependencies like @google/generative-ai.
3. Run npm run compile to build the extension.
4. Launch the extension with F5.
5. Open Settings (Cmd+,), search for "Upscale", and paste your API key.

📂 Project Structure

├── resources/
│   └── hono-docs.md      # Framework-specific standards
├── src/
│   ├── extension.ts      # Command registration & activation
│   ├── lineContext.ts    # Editor context & selection logic
│   ├── llm.ts            # Gemini API integration & prompting
│   ├── docs.ts           # Local documentation provider
│   └── types.ts          # Shared TypeScript interfaces
└── package.json          # Extension configuration & metadata