https://github.com/hardcodedev777/unitycodelama
Unity Edtior tool to analyze your scripts via Ollama
https://github.com/hardcodedev777/unitycodelama
csharp ollama unity unity-editor unity-package unity-scripts
Last synced: 9 months ago
JSON representation
Unity Edtior tool to analyze your scripts via Ollama
- Host: GitHub
- URL: https://github.com/hardcodedev777/unitycodelama
- Owner: HardCodeDev777
- License: mit
- Created: 2025-05-10T20:14:33.000Z (10 months ago)
- Default Branch: main
- Last Pushed: 2025-05-22T16:04:11.000Z (10 months ago)
- Last Synced: 2025-06-02T03:19:45.392Z (10 months ago)
- Topics: csharp, ollama, unity, unity-editor, unity-package, unity-scripts
- Language: C#
- Homepage:
- Size: 2.04 MB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# 🦙 UnityCodeLama — Unity Editor Tool
Listed in [Ollama Community Integrations](https://github.com/ollama/ollama?tab=readme-ov-file#community-integrations)
> 🔍 Analyze your Unity scripts using large language models directly from the Editor — powered by [Ollama](https://ollama.com).
---
## 🚀 Overview
**UnityCodeLama** is a powerful editor extension for Unity that allows you to send your C# scripts and prompts to an LLM (e.g. Qwen, Gemini, Deepseek, etc.) via Ollama — without ever leaving the editor.
It's useful for:
- 🔍 Code explanations
- 🐛 Bug hunting
- ✨ Suggestions & improvements
---
## 📦 Installation
1. Download the latest .unitypackage file from the Releases page.
2. Drag & drop it into your Unity project window.
3. Unity will automatically compile the editor extension.
No additional setup required.
The `Plugins` folder includes several `.dll` files required for integration.
---
## 💻 Usage
Open the tool from the Unity menu:
`HardCodeDev → UnityCodeLama`
Once opened, you’ll see the following window:
---
## 🧰 Editor Window Features
### 📄 Scripts
| Setting | Description |
|------------------------|-------------|
| **Add script path** | Adds a new object field in the window to select a script. |
| **Script file** | Drag & drop or select a `.cs` file to include. |
| **Add script to prompt** | Adds the selected script to the prompt queue. |
| **Clear script list** | Removes all scripts from the queue. |
---
### ⚙️ Properties
| Setting | Description |
|--------------------|-------------|
| **Model name** | Name of the Ollama model to use (e.g. `qwen2.5:3b`). |
| **Clear ** | Removes `` blocks from models that produce internal thoughts (for cleaner output). |
| **Prompt profile** | Selects the type of analysis to perform (Explain, Find bugs, Improve). |
---
### 💬 Prompt
| Setting | Description |
|---------------------|-------------|
| **Prompt** | The custom message to send to the model (e.g. questions, requests). |
| **Send to Ollama** | Sends your prompt, script(s), and profile to Ollama. |
See [How prompts are sent](#how-prompts-are-sent) for more details.
---
### 🤖 Response
| Setting | Description |
|---------------------|-------------|
| **Response** | The LLM-generated reply (read-only). |
| **Copy response** | Copies the reply to your clipboard. |
| **Clear response** | Clears the response field. |
---
> [!NOTE]
> This is an **Editor-only tool**. It works entirely within Unity Editor and doesn't affect runtime or builds.
---
## 📤 How Prompts Are Sent
Say you have a script like:
```csharp
using UnityEngine;
namespace HardCodeDev.Examples
{
public class Test : MonoBehaviour
{
private int _count;
private void Count() => _count++;
}
}
```
You write a prompt like:
```
I made this script in 10 seconds!
```
With profile "Explain", the full message sent to the model will be:
```
Explain what this code does.
I made this script in 10 seconds! Here's the script(s):
// Script starts
using UnityEngine;
namespace HardCodeDev.Examples
{
public class Test : MonoBehaviour
{
private int _count;
private void Count() => _count++;
}
}
// Script ends
```
---
### ⚙️ Code Behind (Simplified)
```csharp
var promptInstruction = _promptOptions[_selectedPromptIndex] switch
{
"Explain" => "Explain what this code does.",
"Find bugs" => "Find any bugs or potential issues in this code.",
"Improve" => "Suggest improvements to this code.",
_ => ""
};
var fullPrompt = $"{promptInstruction}
{_prompt} Here's the script(s): {scriptInfo}";
```
The scripts are read using:
```csharp
scriptInfo += ScriptReader.ReadScript(path);
```
With `ScriptReader` defined as:
```csharp
public static class ScriptReader
{
public static string ReadScript(string relativePath)
{
if (relativePath.StartsWith("/"))
relativePath = relativePath.Substring(1);
var fullPath = Path.Combine(Application.dataPath, relativePath);
var code = "\n// Script starts \n";
code += File.ReadAllText(fullPath);
code += "\n// Script ends \n";
return code;
}
}
```
---
## 🛠 TODO
- [ ] Review which `.dll` files in the `Plugins` folder are actually required and remove the unnecessary ones.
---
## 📄 License
This project is licensed under the **MIT License**.
See the [LICENSE](LICENSE) file for full terms.
---
## 👨💻 Author
**HardCodeDev**
- [GitHub](https://github.com/HardCodeDev777)
- [Itch.io](https://hardcodedev.itch.io/)
---
> 💬 Got feedback, found a bug, or want to contribute? Open an issue or fork the repo on GitHub!