Treelyze

Treelyze is a powerful CLI tool for analyzing directory structures and optionally summarizing file contents using Language Models (LLMs). It provides an easy-to-use interface for exploring and understanding complex directory hierarchies.

Features

• Analyze directory structures with customizable depth
• Exclude specific directories from analysis
• Summarize file contents using LLMs (optional)
• Configurable settings for model and prompt customization
• Output results to file or console

Installation

You can install Treelyze using pip:

pip install treelyze

Usage

Basic Usage

To analyze the current directory:

treelyze run

To analyze a specific directory:

treelyze run /path/to/directory

Options

• --max-depth, -d: Set the maximum depth for directory traversal
• --exclude, -e: Specify directories to exclude (can be used multiple times)
• --summarize, -s: Enable file content summarization using LLM
• --output, -o: Specify an output file for the analysis results
• --model: Override the default language model for summarization
• --prompt: Override the default summarization prompt

Examples

Analyze a directory with a maximum depth of 3, excluding 'node_modules':

treelyze run /path/to/directory --max-depth 3 --exclude node_modules

Analyze and summarize files, outputting results to a file:

treelyze run /path/to/directory --summarize --output results.txt

Configuration

You can manage Treelyze configuration using the config subcommand:

# Show current configuration
treelyze config --show

# Set a configuration value
treelyze config --set model "gpt-3.5-turbo"
treelyze config --set prompt "Summarize this file concisely:"

Note: The configuration file is located at ~/.treelyze/config.yaml in your home folder.

Contributing

Contributions to Treelyze are welcome! Please feel free to submit a Pull Request.

License

This project is licensed under the MIT License - see the LICENSE file for details.