Local AI Coder

A VS Code extension that uses local AI models (via Ollama) to update project code.

Features

• Supports Ollama local AI model server
• Configurable API URL, model name, and timeout settings
• Interactive webview chat interface
• Commands for activation, help, configuration, and exit
• Code operation commands: update, explain, and fix selected code
• Ability to process code changes from AI responses

Requirements

• VS Code 1.75.0 or higher
• Ollama installed and running locally
• A supported model (e.g., qwen3-coder, llama2, etc.)

Installation

1. Clone this repository
2. Run npm install to install dependencies
3. Update copy-assets in package.json against your OS (copy for windows cp for linux)
4. Run npm run compile to build the extension
5. Press F5 to launch the extension in debug mode

Configuration

The extension can be configured via VS Code settings:

• ollama-coder.apiBaseUrl: API base URL for Ollama (default: http://localhost:11434/v1)
• ollama-coder.modelName: AI model to use (default: qwen3-coder:30b)
• ollama-coder.apiKey: API key for Ollama (default: ollama)
• ollama-coder.timeout: Request timeout in seconds (default: 600)

Usage

Basic Chat

1. Activate the extension by running the Activate Local AI Coder command after pressing Ctrl+Shift+P
2. A webview panel will open with the Local AI Coder interface
3. Use the buttons in the panel to:
   • Help - Show available commands and usage information
   • Configure - Open configuration settings
   • Exit - Close the panel
4. Use the chat input box to send messages to the AI:
   • Type your message in the input field
   • Click "Send" button or press Enter
   • The AI response will appear in the messages area
5. Configure your AI model settings in VS Code settings before using the chat feature

Code Operations

1. Update Code: Select code in the editor, then run Update Selected Code command. Enter instructions for the update when prompted.

2. Explain Code: Select code in the editor, then run Explain Selected Code command. The AI will provide a detailed explanation of the code.

3. Fix Code: Select code in the editor, then run Fix Selected Code command. Describe the problem when prompted, and the AI will fix the code.

Commands

• local-ai-coder.activate: Activate the extension and open webview panel
• local-ai-coder.help: Show help message
• local-ai-coder.configure: Open configuration settings
• local-ai-coder.exit: Deactivate the extension
• local-ai-coder.updateCode: Update selected code based on instructions
• local-ai-coder.explainCode: Explain selected code
• local-ai-coder.fixCode: Fix selected code based on problem description

Example

Basic Chat Example

1. Activate the extension by running Activate Local AI Coder command
2. The webview panel will open automatically
3. Click "Configure" button to open settings and verify your AI model configuration
4. In the chat input box, type a message like "Refactor this function to be more efficient"
5. Click "Send" button or press Enter
6. The AI will respond with suggestions and code changes in the messages area

Code Update Example

1. Select some code in the editor
2. Open the command palette (Ctrl+Shift+P)
3. Run Update Selected Code command
4. Enter instructions like "Make this function more efficient"
5. The AI will provide updated code based on your instructions

Supported Models

• qwen3-coder (recommended)
• llama2
• Any other model supported by Ollama

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

License

MIT