Yalla - Project to LLM Aggregator 📁

a Python script to prepare your project for LLM interactions. Yalla generates a single-file view, including the directory structure, helping LLMs in understanding the project's context and file structure for more accurate responses.

How doe it work

Yalla walks through your project directory and:

1. Creates a visual tree of your project structure
2. Combines the contents of all text files into a single output file
3. Follows project's ignore patterns (similar to .gitignore)

Basic Usage

# creates output.txt in current directory
python yalla.py

# Specify custom output file
python yalla.py -o my_output.txt # or --output

# Default behavior (creates output.txt ignoring .yallaignore patterns)
python yalla.py

# Include .gitignore patterns in addition to .yallaignore
python yalla.py -g  # or --gitignore

# Help
python yalla.py -h ## or --help

*It is reccomended to add yalla.py directory to PATH so that you can run simply *yalla* from wherever you want.

Ignore Patterns

By default, Yalla looks for a .yallaignore file in your project root. The syntax is similar to .gitignore:

• Each line represents one pattern
• Lines starting with # are comments
• Empty lines are skipped
• Add / at the end of a pattern to match directories only
• Supports standard glob patterns (* and ?)

Default ignored files:

• Any hidden folder in your OS
• .yallaignore
• output.txt

Requirements

• Python 3.x
• std only, no other libraries needed

Output Format

The generated file includes:

Directory Structure:
  folder1/
    - file1.txt
    - file2.py
  folder2/
    subfolder/
      - file3.js

# folder1/file1.txt
[content of file1.txt]

# folder1/file2.py
[content of file2.py]
...

Why "Yalla"? 😎

Because sometimes you just need to get all your code together quickly - yalla! (Hebrew for "let's go" or "hurry up")