ai-digest

ai-digest is a CLI tool that aggregates your codebase into a single Markdown file. It is designed for use with Claude Projects or custom ChatGPTs, providing a summarized, clean, and easily shareable format of your project's code.

Installation

To install ai-digest globally on your system, use the following command:

Install via Cargo

If you have Rust and Cargo installed, you can install ai-digest globally using this command:

cargo install ai-digest

Once installed, the ai-digest command will be available from anywhere on your system.

Usage

Once installed, you can use ai-digest from any project or directory by running:

ai-digest --input <directory> --output <output-file> [options]

Options

• --input or -i: Specify the input directory to scan for files. Default is the current directory (.)
• --output or -o: Specify the output file for the generated Markdown. Default is codebase.md
• --no-default-ignores: Disable default ignore patterns (target, node_modules, .git)
• --whitespace-removal: Enable whitespace removal from file contents
• --show-output-files: Display a list of files included in the output
• --ignore-file: Specify a custom ignore file. Default is .aidigestignore

Example

ai-digest --input ./path/to/project --output output.md --whitespace-removal --show-output-files

This will generate a Markdown file named output.md, summarizing the project in the ./path/to/project directory. The output will have whitespace removed, and all included files will be listed in the output.

Development

To contribute to ai-digest or modify it for your needs, clone the repository and build it locally:

git clone https://github.com/adityarb2003/ai-digest.git
cd ai-digest
cargo build

Running Tests

To run the unit tests:

cargo test

Publish to Crates.io

To publish ai-digest to Crates.io:

1. Log in to your Crates.io account:

cargo login

2. Publish the crate:

cargo publish

This will upload the crate to Crates.io, making it available for global installation via cargo install ai-digest.

License

Distributed under the MIT License. See LICENSE for more information.