A Python automation tool that helps fill out timesheets by simulating mouse clicks, keyboard input, and scrolling actions based on pre-defined commands.

• Automated Button Clicking: Finds and clicks buttons on the screen using image recognition
• Text Input Automation: Automatically types text from text files
• Mouse Scroll Simulation: Simulates mouse wheel scrolling
• Action Recording: Includes a recorder utility to create automation sequences

• bot_timesheet.py: Main script that executes the automated actions
• recorder.py: Utility to record mouse clicks, keyboard input, and scrolling actions
• comandos/: Directory containing all the command files (images, text files) used by the bot

1. Button Commands: Stored as .bmp files in the format <order>_boton_x<x_offset>_y<y_offset>_t<delay>.bmp

   • These are button images the bot will look for on screen
   • x_offset and y_offset define the click position relative to the button's center
   • t specifies how long to wait after clicking

2. Text Commands: Stored as .txt files in the format <order>_texto_t<delay>.txt

   • These contain text the bot will type
   • Before typing, it selects all existing text (Ctrl+A) and deletes it

3. Scroll Commands: Stored as .txt files in the format <order>_scroll_t<delay>.txt

   • These contain the scroll amount (positive for up, negative for down)

• Python 3.6+
• PyAutoGUI
• PIL (Pillow)
• keyboard
• pynput
• OpenCV-Python (for image recognition confidence)

# Clone the repository
git clone https://github.com/StructuralWizard/bot_timesheet.git
cd bot_timesheet

# Install dependencies
pip install -r requirements.txt

python bot_timesheet.py

The bot will:

1. Wait 3 seconds (to allow you to position your cursor)
2. Execute all commands in the comandos directory in alphabetical order

python recorder.py

The recorder allows you to:

• Press F8 to record the position where you want to click
• Press F9 to mark the top-left and then bottom-right corners of the button
• Type to record keyboard input
• Use the scroll wheel to record scrolling
• Press F10 to stop recording

• The image recognition uses OpenCV for reliability. If you encounter issues, adjust the confidence parameter in the code.
• For button clicks, make sure the button images are clear and distinctive.
• The bot runs actions in alphabetical order of the filenames, so naming conventions are important.

This project is open source and available under the MIT License.