dir2chat 0.1.0a0

Generate an OpenAI Chat Completions-compatible JSON from all UTF-8 plaintext files in a directory.

dir2chat

Generate an OpenAI Chat Completions-compatible JSON from all UTF-8 plaintext files in a directory.

Use Cases

This tool is designed to help with a variety of tasks involving directory contents and large-language models, such as:

• Codebase Summarization for LLMs
  Preparing a repository as an input for Chat Completions-compatible models, making it easy to "upload" a codebase as a conversational context.

• Automated Project Documentation
  Quickly gather all plaintext documentation files (like README.md, docs/, etc.) and their contents for further summarization, analysis, or indexing.

Installation

pip install dir2chat

Usage

python -m dir2chat <directory> [options]

Options:

• -o, --output: Output JSON file (default: - for stdout).
• --max-file-size: Max size for files in bytes (default: 65536, or 64 KB).

Example:

python -m dir2chat path/to/my/project -o chat.json --max-file-size 32768

Output

The output is a JSON array of messages suitable for OpenAI Chat models, in the structure:

[
  {
    "role": "user",
    "content": "- README.md\n- src/\n  - main.py\n  - utils.py"
  },
  {
    "role": "user",
    "content": "README.md"
  },
  {
    "role": "user",
    "content": "<contents of README.md>"
  },
  ...
]

1. First message: The tree view of the directory.
2. Subsequent pairs: For each included file, a message with the file's relative path, then another with its content.

Only UTF-8-encoded, plaintext files below the size limit are included. Skipped files are reported to stderr.

Contributing

Contributions are welcome! Please submit pull requests or open issues on the GitHub repository.

License

This project is licensed under the MIT License.