🦙 Ollama Code Fixer - AI Coding Assistant

Comprehensive AI-powered coding assistant using local Ollama models in VS Code.

🌟 Features

🔧 Code Operations

Fix Code - Automatic error and bug fixing
Optimize Code - Performance and readability optimization
Explain Code - Detailed explanations of code logic
Add Comments - Generate comments and documentation
Generate Tests - Create unit tests with edge case coverage
Refactor Code - Improve structure and architecture
Security Check - Analyze security vulnerabilities
Generate Code - Create code from descriptions
Translate Code - Convert between programming languages

🚀 Ollama Management

Auto-start Server - Automatic Ollama startup when needed
Model Selection - Quick switching between installed models
Model Installation - Download and install new AI models
Status Monitoring - Real-time API status tracking

🎯 Smart Features

Auto Code Insertion - Configurable change application
Preview Mode - Preview changes before applying
Flexible Positioning - Replace, insert above/below, or new file
Backup Creation - Automatic original code backup
Intelligent Chat - Full-featured AI coding assistant

🛠 Installation

Prerequisites

Ollama - Download and install
VS Code version 1.85.0 or higher

Extension Installation

Open VS Code
Go to Extensions (Ctrl+Shift+X)
Search for "Ollama Code Fixer"
Click Install

Ollama Setup

# Install a model (e.g.)
ollama pull codellama:7b

# Start the server
ollama serve

🎮 Usage

Select code in the editor
Right-click → choose operation:
- 🔧 Fix Code
- ⚡ Optimize Code
- 📝 Explain Code
- 💬 Add Comments
- 🧪 Generate Tests
- 🔄 Refactor Code
- 🔒 Security Check

Command Palette

Press Ctrl+Shift+P
Type "Ollama" to view all commands

Click Ollama icon in Activity Bar
Use the tool panel:
- 🚀 Start Ollama - start server
- 🎯 Select Model - choose model
- 📥 Install Model - install new model
- 💬 Open AI Chat - open chat

Code Generation

Place cursor where needed
Ctrl+Shift+P → "Generate Code"
Describe what you want to create
AI generates code from description

⚙️ Configuration

Basic Settings

{
  "ollamaCodeFixer.modelName": "codellama:7b",
  "ollamaCodeFixer.ollamaApiUrl": "http://localhost:11434",
  "ollamaCodeFixer.language": "en",
  "ollamaCodeFixer.autoApplyChanges": false,
  "ollamaCodeFixer.showPreviewBeforeApply": true
}

Behavior Settings

autoApplyChanges - Auto-apply without confirmation
insertPosition - Where to insert code: replace, above, below, newFile
backupOriginalCode - Create backup copies
autoStartOllama - Auto-start Ollama server

Model Settings

temperature (0.0-2.0) - Response creativity
topP (0.0-1.0) - Nucleus sampling
topK (1-100) - Top-k sampling
maxTokens - Maximum response length
contextLength - Context size

🌐 Multilingual Support

Supported interface languages:

🇺🇸 English (default)
🇷🇺 Russian
🇺🇦 Ukrainian
🇪🇸 Spanish

{
  "ollamaCodeFixer.language": "en"  // en, ru, uk, es
}

Change language: Ctrl+Shift+P → "🌐 Change Language"

🔧 Supported Programming Languages

Web: JavaScript, TypeScript, HTML, CSS, SCSS
Backend: Python, Java, C#, Go, Rust, PHP, Ruby
Mobile: Swift, Kotlin, Dart
Systems: C, C++, Scala
Data: SQL, R, Julia
DevOps: Shell, PowerShell, Docker, YAML
And many more...

🎯 Recommended Models

For Programming

codellama:7b - Optimal balance of speed and quality
codellama:13b - Higher quality
codegemma:7b - Fast code processing

General Purpose

llama2:7b - Good overall quality
mistral:7b - Fast and efficient
gemma:7b - Balanced performance

📊 Usage Examples

Error Fixing

# Select problematic code
def fibonacci(n):
    if n <= 1:
        return n
    return fibonacci(n-1) + fibonacci(n-2)  # Inefficient

# Right-click → Fix Code
# Get optimized version with memoization

Test Generation

// Select function
function validateEmail(email) {
    const re = /^[^\s@]+@[^\s@]+\.[^\s@]+$/;
    return re.test(email);
}

// Right-click → Generate Tests
// Get comprehensive test suite

Code Explanation

-- Select complex query
SELECT u.name, COUNT(o.id) as order_count,
       AVG(o.total) as avg_order_value
FROM users u
LEFT JOIN orders o ON u.id = o.user_id
WHERE u.created_at > '2024-01-01'
GROUP BY u.id, u.name
HAVING COUNT(o.id) > 5;

-- Right-click → Explain Code
-- Get detailed logic explanation

🔄 Workflow

Select code or place cursor
Choose operation via context menu or Command Palette
Preview result (if preview is enabled)
Apply changes or save to new file
Use chat for additional questions

🛡️ Security

Local Processing - Your code never leaves your machine
No Internet Requests - Everything works through local Ollama
Backup Creation - Automatic original code backup
Preview Mode - Control before applying changes

🎨 User Interface

Status Bar

$(check) Ollama: Active - Server running
$(error) Ollama: Offline - Server unavailable
Click to check status

Complete toolset with icons:

🔧 Code fixing
⚡ Optimization
📝 Explanations
💬 Comments
🧪 Tests
🔄 Refactoring
🔒 Security
✨ Generation
💬 AI Chat

🔍 Debugging

Logging

{
  "ollamaCodeFixer.logLevel": "debug"
}

Log levels:

error - Errors only
warn - Warnings
info - Information (default)
debug - Detailed debugging

Status Check

Command Palette → "Ollama: Check API Status"
Status Bar → click Ollama status
Output Panel → "Ollama Code Fixer"

🤝 Contributing

Fork the project
Create feature branch
Make changes
Add tests
Create Pull Request

📋 TODO

[ ] CodeLens support
[ ] Git integration
[ ] Full project analysis
[ ] Custom prompts
[ ] Framework-specific plugins
[ ] Cloud model support

📝 Changelog

v0.3.0 - 2024-12-19

🚀 New Features

Full AI agent with 9 different code operations
Multilingual support - 4 interface languages (EN, RU, UK, ES)
Ollama management - auto-start, model selection and installation
Automatic model installation with real-time progress bar
Flexible insertion settings - replace, above, below or new file
Context menu with full operation set for selected code
Preview mode for changes before applying
Backup creation of original code to clipboard
Auto-start Ollama when server unavailable

🎨 Interface Improvements

New icons for all operations
Enhanced sidebar with extended functionality
Informative status bar with clickable status
Welcome message for new users

⚙️ Settings

Extended model parameters (temperature, top_p, top_k)
Behavior settings (auto-apply, preview, position)
Model management (ollama path, preferred models)

📞 Support & Contacts

🐛 GitHub Issues: Report a Problem
💬 Discussions: Ideas & Suggestions
📖 Documentation: README on GitHub
⭐ Star on GitHub: Support the Project

📄 License

MIT License - see LICENSE file for details.

🔗 Useful Links

📦 VS Code Marketplace: Download Extension
🦙 Ollama: Official Website
📚 Ollama Models: Model Library
💻 VS Code API: Developer Documentation

🤝 Contribute

This project is open for contributions! If you have ideas for improvement:

Fork this repository
Create feature branch (git checkout -b feature/AmazingFeature)
Commit your changes (git commit -m 'Add some AmazingFeature')
Push to branch (git push origin feature/AmazingFeature)
Open Pull Request

Made with ❤️ for developers using local AI models

Ollama Code Fixer - AI Coding Assistant

Ruslan