v0.1.0

Release v0.1.0 - Initial Release

Public release of Cumul!

cumul is a simple yet powerful command-line utility, written in Zig, designed to help developers and researchers consolidate source code and text files into a single file.

This is especially useful for providing large amounts of context to Large Language Models (LLMs) for analysis, review, or training.

This initial version lays the foundation for a fast and reliable developer tool that streamlines your workflow with AI.

✨ Key Features

• File Concatenation: Recursively scans a directory and combines all valid text files into a single, well-formatted ...-cumul.txt file.
• Smart .gitignore Integration: Automatically respects the rules in your project's .gitignore file, ensuring that ignored files and build artifacts are not included in the output.
• Intelligent Filtering: Skips binary files (like images), hidden dotfiles, and other non-essential files to keep the output clean and relevant.
• Simple & Fast CLI: A straightforward command-line interface (cm) that gets the job done quickly.
• Customizable Output: Use the -p or --prefix flag to add a custom prefix to the output filename, making it easier to manage multiple contexts.

Usage Examples

1. Run in the current directory:

   cm

   This will generate a file like my-project-cumul.txt.

2. Specify a target directory:

   cm ../path/to/another-project

3. Add a prefix to the output file:

   cm -p feature-branch .
   ```    *This will generate a file named `feature-branch-my-project-cumul.txt`.*

Binaries Included in this Release

• cumul-linux-x86_64.tar.gz
• cumul-linux-aarch64.tar.gz
• cumul-macos-x86_64.tar.gz
• cumul-macos-aarch64.tar.gz

Thank you for checking out the project.

Contributions are welcome.