Codexpilot is a VS Code extension that integrates Google's Gemini AI to provide intelligent code assistance with codebase context. Simply chat with Gemini and reference your project files using '@' mentions to get contextually relevant help with your code.
Features
Google Gemini Integration: Powered by Google's advanced Gemini 1.5 Pro model for intelligent code understanding and generation
Automatic Context Tracking: The currently active file is automatically added to the chat context with a toggle to control inclusion
Codebase Context via '@' Mentions: Easily reference additional files in your workspace by typing '@' to provide context to the AI
Hybrid Context Model: Combines auto-tracked current file with persistent manually added files for comprehensive context
File Search: Quick access to recent files or search your entire workspace
Streaming Responses: See AI responses in real-time as they're generated
Stop Generation: Cancel an in-progress AI response at any time with the Stop button
Input Validation: Character limit with visual feedback prevents oversized queries
Rich Markdown Rendering: Beautiful formatting with syntax highlighting for code blocks
Code Actions: Copy code snippets to clipboard or insert them directly at your cursor position
Chat Management: Start new conversations with a single click
Secure API Key Storage: Your Gemini API key is securely stored using VS Code's built-in secret storage
Requirements
Visual Studio Code v1.74.0 or higher
Google AI Studio API Key for Gemini
Installation
Open VS Code
Go to Extensions (Ctrl+Shift+X)
Search for "Codexpilot"
Click Install
Getting Started
Obtaining a Gemini API Key
To use Codexpilot, you'll need a Google AI Studio API key: