ZeroIndex

ZeroIndex is a high-performance, visual developer tool for VS Code that makes code comprehension instantaneous. Hover over any class, method, or function declaration, and ZeroIndex will immediately present a clean, AI-generated explanation of its purpose, parameters, and return details, accompanied by an interactive learning log in your sidebar.

🚀 Key Features

• Instant Hover Explanations: View context-rich documentation right inside your editor as you read code. No more jumping across files to understand a class or method's logic.
• Flexible AI Provider Engine: Seamlessly connect to your preferred model provider:
  • Google Gemini (Gemini 2.5 Flash, 2.5 Pro, 2.0, 1.5)
  • OpenAI (GPT-4o, GPT-4o-mini, GPT-3.5)
  • Anthropic (Claude 3.5 Sonnet, Claude 3.5 Haiku)
  • Ollama (100% offline, local models like llama3, codellama, etc.)
• Interactive Exploration Sidebar: A dedicated learning log in your primary sidebar lists the code constructs you explored. Search, filter by starred elements, delete history entries, and jump back to the original definition with one click.
• Intelligent Local Caching: Automatically caches explanations locally, ensuring sub-millisecond retrieval speeds and minimizing API token consumption.
• Granular Hover Toggles: Disable or enable hover descriptions for classes and methods independently via user settings or quick tooltip links to keep your workspace clean.
• Markdown Export: Export your day's learning log to a clean, formatted Markdown document—ideal for daily standups, team knowledge sharing, or code reviews.

🛠️ Getting Started

1. Installation

Install the extension from the VS Code Marketplace.

2. Configure Your AI Connection

ZeroIndex launches an onboarding wizard upon initial activation. You can open it anytime:

1. Open the VS Code Command Palette (Cmd+Shift+P on macOS / Ctrl+Shift+P on Windows/Linux).
2. Run the command ZeroIndex: Open Settings Dashboard.
3. Choose your AI provider, paste your credentials (or local endpoint), and select Test & Save.

⌨️ Extension Commands

ZeroIndex contributes the following commands to the VS Code Command Palette:

⚙️ Configuration Reference

Fine-tune ZeroIndex configuration directly in your user settings:

• zeroindex.provider: Active AI provider (openai, anthropic, gemini, ollama).
• zeroindex.openai.model: Model identifier to use with OpenAI (e.g. gpt-4o-mini).
• zeroindex.anthropic.model: Model identifier for Anthropic (e.g. claude-3-5-sonnet-20241022).
• zeroindex.gemini.model: Model identifier for Google Gemini (e.g. gemini-2.5-flash).
• zeroindex.ollama.model: Local model name running in Ollama (e.g. llama3).
• zeroindex.ollama.endpoint: HTTP endpoint of your Ollama service (defaults to http://localhost:11434).
• zeroindex.explanationLength: Toggle between concise (1-2 sentences) and detailed explanations.
• zeroindex.triggerMode: Choose to fetch explanations on hover, command invocation, or both.
• zeroindex.cacheEnabled: Enable or disable local explanation caching.
• zeroindex.explainClasses: Show/hide hovers when cursor is placed on a class symbol.
• zeroindex.explainMethods: Show/hide hovers when cursor is placed on a method/function symbol.

📄 License

This project is licensed under the MIT License.