ollama-code

Cost-effective AI code assistant - 100% local, $0 cost alternative to Claude Code

Features

• 🚀 100% Local - Run everything on your machine, no cloud dependencies
• 💰 Zero Cost - Use free local LLMs via Ollama
• 🛠️ Full-Featured - File operations, code execution, SQLite, and more
• 🎯 Production Ready - Robust error handling, testing, and performance optimizations
• 🔌 Plugin System - Extend with custom tools and commands
• 🤖 MCP Server - Model Context Protocol support for advanced integrations

Installation

npm install -g ollama-code

Prerequisites

• Node.js >= 18.0.0
• Ollama installed and running locally
• A model installed in Ollama (e.g., ollama pull qwen2.5-coder)

Quick Start

# Start the interactive CLI
ollama-code

# Run with a specific model
ollama-code --model qwen2.5-coder

# Start as MCP server
ollama-code-mcp

Configuration

Create a .ollama-code.yaml file in your project root:

model: qwen2.5-coder
temperature: 0.7
max_tokens: 4096
tools:
  enabled: true
  parallel: true

Available Tools

• File Operations: Read, write, edit files
• Code Execution: Run bash commands and scripts
• Search: Advanced grep and file search
• SQLite: Database operations
• Web Fetch: Retrieve and analyze web content
• Plugin System: Add custom tools and commands

CLI Commands

• /help - Show available commands
• /clear - Clear conversation history
• /model <name> - Switch to a different model
• /tools - List available tools
• /exit - Exit the CLI

Development

# Clone the repository
git clone https://github.com/yourusername/ollama-code.git
cd ollama-code

# Install dependencies
npm install

# Build the project
npm run build

# Run tests
npm test

# Start in development mode
npm run dev

Architecture

• TypeScript - Full type safety
• Zod Validation - Runtime type checking
• Plugin System - Extensible architecture
• Error Handling - Comprehensive error recovery
• Performance - Optimized algorithms and memory management
• Testing - 80+ tests with Vitest

License

MIT

Contributing

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

Support

For issues and feature requests, please visit GitHub Issues.