A CLI for talking to ChatGPT, saving conversation logs, and turning code blocks (language agnostic) from it's responses into executable shell commands.

1. Install the python dependencies:

   pip install -r requirements.txt
   

2. Create an API key on the OpenAI developer portal.

3. Replace YOUR_API_KEY in the .env file with your OpenAI API key.
   OR
   You can run the following command: (however this does not get persisted on reboot)

   export OPENAI_API_KEY=<your-api-key>
   

4. Run the setup script. This will add the gptea alias for your script in .bashrc for easy execution in any directory, and add local directories (generated-scripts/ and conversations/) for saving ChatGPT: generated files or conversations.

   chmod +x setup.sh && source setup.sh
   

5. You are good to go! 🤠 Run gptea to get started.

1. _ : no input will display the list of commands.
2. ask : ask ChatGPT: a question without saving any files.
3. convo : have a continuous conversation with ChatGPT:.
4. lang <lang-name> : ask ChatGPT: to create a bash script that uses the input language (<lang-name>) under the hood.
   Available <lang-name> values:
   python, bash, javascript, typescript, java, c, c++, c#, rust, swift, go, ruby, php