Code Buddy - Your AI Co-Developer

Code Buddy generates Unity code that already understands your project - your existing classes, methods, and architecture - without you having to explain them every time.

Stop copying your code into ChatGPT or fighting with generic AI assistants that don't know your codebase. Code Buddy analyzes your project locally and generates scripts that integrate immediately, reducing manual refactoring from hours to minutes.


New in Version 3: Manipulate your scenes and GameObjects in the same chat. Place prefabs, configure components, and assign references using natural language - perfect for rapid prototyping.


🚀 Project-aware code generation

Generate C# scripts and shaders that use your existing classes correctly. Code Buddy scans your project locally to understand your architecture, so the AI produces code that fits your project instead of generic examples.

Works with OpenAI, Claude, Google Gemini, DeepSeek, or local models via Ollama for completely offline use.


🤖 Scene & GameObject automation

Skip the clicking. Use text prompts to:

  • Place and arrange prefabs in your scene
  • Configure component settings automatically
  • Assign GameObject references and dependencies
  • Prototype level layouts in seconds

Perfect for repetitive editor tasks that waste your time.


💬 Integrated code chat

Discuss your code directly in Unity. Ask for refactoring suggestions, explanations of complex logic, or alternative approaches. Useful when inheriting unfamiliar code or planning new features.


🏷️ No subscription required

One-time purchase. Bring your own API key for OpenAI, Claude, Gemini, or DeepSeek—you control costs with pay-as-you-go pricing. Or install Ollama to run local models completely free and offline.


🎮 Unity-first workflow

Everything happens in the Unity Editor. No context-switching to web browsers or external tools. Generated code follows Unity C# conventions automatically.


🎓 Accessible prompting

No prompt engineering required. Use natural language for standard tasks. For complex problems, iterate in the chat to refine your approach before generating code.


🌍 Proxy support

Configure Code Buddy to route API requests through a proxy server if your region blocks access to LLM providers.


🛡 Privacy first

Your code never leaves your machine. Analysis happens locally. Only your prompt and minimal context are sent to your chosen AI provider (or nothing at all with Ollama). API keys are encrypted and stored locally.

Updated Context System: Better support for larger projects, improved generation accuracy, and lower API token usage.


⚠️ IMPORTANT ⚠️

Requires user setup of one of the following:

1. OpenAI API Key (does not work with ChatGPT Plus subscription).

2. DeepSeek API Key.

4. Gemini API Key.

5. Claude AI API Key.

3. Ollama installed locally


Join Discord to see the roadmap.


JOIN DISCORD | DOCUMENTATION | YOUTUBE