Metadata-Version: 2.2
Name: bourguibagpt
Version: 1.7.7
Summary: Tunisian Shell Command Assistant
Home-page: https://github.com/AzizBahloul/PromptToCommand
Author: Si Aziz Bahloul
Author-email: azizbahloul3@gmail.com
License: MIT
Classifier: Programming Language :: Python :: 3
Classifier: License :: OSI Approved :: MIT License
Classifier: Operating System :: OS Independent
Classifier: Development Status :: 4 - Beta
Classifier: Intended Audience :: Developers
Classifier: Natural Language :: English
Classifier: Topic :: Software Development :: Libraries :: Python Modules
Requires-Python: >=3.6
Description-Content-Type: text/markdown
Requires-Dist: rich==13.6.0
Requires-Dist: requests>=2.25.0
Requires-Dist: torch>=1.13.0
Requires-Dist: transformers>=4.30.0
Requires-Dist: bitsandbytes>=0.39.0
Requires-Dist: accelerate>=0.20.0
Requires-Dist: psutil>=5.9.0
Requires-Dist: tqdm>=4.65.0
Requires-Dist: sentencepiece>=0.1.99
Requires-Dist: huggingface-hub>=0.16.4
Requires-Dist: inquirer>=3.4.0

---

**BourguibaGPT 🇹🇳**  

Your AI-Powered Tunisian Shell Command Assistant  

Python Version: 3.6+  
License: MIT  
PyPI Version: [Check Here](https://pypi.org/project/bourguibagpt/)  

BourguibaGPT is an intelligent CLI tool that transforms natural language requests into safe shell commands using advanced AI models. Designed with a focus on safety and usability, it's your perfect companion for terminal productivity.  

CLI Demo: [View Image](https://via.placeholder.com/800x400.png?text=BourguibaGPT+Interactive+CLI+Demo)  

---

### Features ✨  

- **Natural Language Understanding** 🇹🇳  
  Describe tasks in plain English, French, or Arabic (Tunisian dialect friendly!).  
- **AI-Powered Command Generation** 🤖  
  Leverages local LLMs via Ollama (default: deepseek-r1:1.5b).  
- **Safety First** 🔒  
  Multi-layer validation against dangerous commands.  
- **Interactive CLI** 💻  
  Rich terminal interface with syntax highlighting and progress indicators.  
- **Command History** 📚  
  Automatic tracking of generated commands with execution status.  
- **Cross-Platform** 🌍  
  Supports Linux, macOS, and Windows Subsystem for Linux.  
- **Smart Execution** ⚡  
  Optional auto-execution with confirmation prompts.  

---

### Installation 📦  

**Step 1: Install Ollama (required)**  

Run the following command in the terminal:  

`curl -fsSL https://ollama.com/install.sh | sh`  
`ollama pull deepseek-r1:1.5b`

**Step 2: Install BourguibaGPT**  

Now install BourguibaGPT with:  

`pip install bourguibagpt`

---

### Quick Start 🚀  

Run the command:  

`bourguibagpt`  

Example interaction:  

```
╔══════════════════════════════════════════════════════════════╗
║              Your Tunisian Shell Command Assistant           ║
╚══════════════════════════════════════════════════════════════╝

🇹🇳 BourguibaGPT → list all json files modified last week
Generated command: find . -name "*.json" -mtime -7

Execute? [yes/no]: yes
✅ Command executed successfully
```  

---

### Advanced Usage 🛠️  

#### Command Line Options  

`bourguibagpt [OPTIONS]`  

- `--model MODEL_NAME` → Specify Ollama model (default: deepseek-r1:1.5b).  
- `--temperature FLOAT` → Control creativity (0.1-1.0, default: 0.7).  
- `--auto-execute` → Auto-execute validated commands.  
- `--history-file PATH` → Custom command history location.  

#### Manage History  

Show last 10 commands:  
`bourguibagpt --history`  

Execute from history:  
`bourguibagpt execute "find . -name *.json -mtime -7"`  

---

### Safety Measures ⚠️  

BourguibaGPT includes multiple protection layers:  

- Pattern matching against dangerous commands (e.g., `rm -rf /`).  
- Command structure validation.  
- Execution confirmation prompts.  
- Restricted special characters handling.  
- Model prompt engineering for safety.  

---

### Roadmap 🗺️  

- [ ] Browser-based GUI  
- [ ] Command explanation mode  
- [ ] Plugin system for custom validations  
- [ ] Shared team history sync  
- [ ] Alternative AI backend support  

---

### Contributing 🤝  

We welcome contributions! Please see our Contribution Guidelines (CONTRIBUTING.md) for details.  

---

### License 📄  

This project is licensed under the MIT License - see the LICENSE file for details.  

---

GitHub Repository: [Visit Here](https://github.com/AzizBahloul/PromptToCommand)  
Report Issues: [Submit Here](https://github.com/AzizBahloul/PromptToCommand/issues)  

---
```

