Ggama is a Python-based local interface to ChatGPT with additional context provided in vector form based on files in a github repository, allowing you to ask it questions about that repository in detail.
- OpenAI access token,
- GitHub access token,
- GitHub repository,
- Python 3.7 or higher,
- eel package (for web api),
- llama-index and llama-index-readers-github,
1- Install requirements, 2- Copy ggama.user.toml-template to ggama.user.toml and edit your own values in, 3- Run app.py which will launch a very simple eel-based chat interface using the specified GitHub repository as a source of data and OpenAI's ChatGPT as an inference engine,
e.g
# prep
$ python -m venv venv
$ . ./venv/bin/activate
# . ./venv/scripts/activate.ps1 for powershell on win/lin/mac
$ pip install -r requirements.txt
# config
$ cp config.user.toml-template config.user.toml
$ $EDITOR config.user.toml # <- edit settings
# run
$ python app.py