HEXEC

An intelligent AI-powered command control system for Windows that enables natural language interaction to execute commands, run code, and manage your system.

Created by: Dequs

---

Features

Core Features:

• Natural Language Control: Communicate with AI in plain English to control your Windows system
• Code Execution: Execute Python code generated by AI with safety confirmations
• Command Execution: Run Windows CMD commands directly from chat
• Chat History: Persistent chat history with unique session identifiers
• Configuration Management: Easy setup and modification of API keys and model selection
• Update System: Built-in update checking and application updates
• Safety Mode: Optional confirmation prompts before executing commands and code
• Color-Coded UI: User-friendly terminal interface with color-coded responses

---

Requirements

• Python 3.7+
• Windows OS (Windows 10/11 recommended)
• Internet Connection (for API calls)
• API Key (Gemini API or compatible service)

---

Installation

1. Clone or Download the Repository

git clone https://github.com/Dequs/HEXEC.git
cd HEXEC

2. Install Dependencies

The application automatically checks for and installs missing dependencies on first run.

3. First Run

start.bat

On first run, you'll be prompted to enter:

• API Key: Your Gemini API key
• API Key Comment: An optional description/label for your API key
• Model: AI model to use (e.g., gemini-2.5-flash, only supported by generativeai from google)
• Always Ask Mode: Whether to confirm before executing commands (y/n)

---

Configuration

Initial Setup

Configuration is stored in config.json. After first run, you can modify settings with the config command in the application.

Update Configuration

While in the application, use:

>>> config

Then follow the prompts to update:

• API Key
• Model selection
• Always Ask mode preference

---

Usage

Starting the Application

python main.py

run start.bat

Chat Interface

1. Select a Chat Session:

   • View available chat histories
   • Start a new chat
   • Or delete all chat histories

2. Interact with AI:

   >>> What is the current time?
   >>> Create a file called test.txt
   >>> List all files in the current directory
   

3. Built-in Commands:

   • exit - Exit the application
   • config - Update configuration
   • cls / clear - Clear the screen
   • history - Reset chat history
   • reset - Same as history

Safety Features

• Always Ask Mode: When enabled, you'll be prompted to confirm before:

  • Executing any Windows CMD commands
  • Running Python code

• Dry Run & Safe Mode: The command executor can validate commands before execution

---

Project Structure

HEXEC/
├── main.py                 # Main application entry point
├── config.json             # Configuration file (auto-generated)
├── prompt.txt              # System prompt for AI behavior
├── logs.log                # Full detailed application logs
├── promptComments.txt      # Prompt for response processing
├── promptThinking.txt      # Prompt for deep thinking
├── VERSION                 # Version information
├── README.md               # This file
├── utils/
│   ├── client.py          # AI client/API communication
│   ├── commands.py        # Command execution logic
│   ├── config.py          # Configuration management
│   └── functions.py       # Utility functions
└── chats/                 # Chat history storage
    └── *.elham            # Individual chat session files

---

How It Works

1. User Input: You provide commands in natural language
2. AI Processing: The AI analyzes your request and generates a response
3. JSON Parsing: Response is parsed as JSON containing command(s) to execute
4. Confirmation: If "Always Ask" is enabled, you approve command execution
5. Execution: Commands are executed and results are captured
6. Feedback: Results are sent back to AI for context-aware responses
7. Display: Final response is displayed to the user

---

Supported Commands

Python Code

• File operations (create, read, write, delete)
• Data processing and calculations
• System information retrieval
• Any Python standard library operations

Windows CMD

• File operations (dir, copy, move, del, etc.)
• Directory management (mkdir, rmdir, etc.)
• System queries
• Text processing
• Network operations

---

API Support

Currently supports:

• Google Gemini API (default)

The application uses the model specified in configuration (e.g., gemini-2.5-flash).

To use a different API, modify the client.py file in the utils/ folder.

---

Troubleshooting

Issue: "Error parsing response JSON"

• The AI response wasn't valid JSON
• Try asking your question again with more specific terms

Issue: "Command execution failed"

• The command may be invalid for your system
• Check that you're using Windows-compatible commands
• Review the command before confirming execution

Issue: "API Key error"

• Verify your API key is correct
• Check your internet connection
• Ensure your API key has the necessary permissions

---

Version Management

Check your current version at application startup. The system automatically checks for updates and offers to apply them.

---

Contributing

Contributions are welcome! Feel free to fork, modify, and submit pull requests.

---

License

Please check the LICENSE file for licensing information.

---

Support

For issues, questions, or suggestions, please open an issue on GitHub.

---

Created with by Dequs