Let's discover Kel in less than 5 minutes.
Kel is your AI assitant in your CLI.
What you'll need
- Python3.6 or above
- respective LLM's API keys
How to get started?
kel repo to your local.
pip install .
You can type this command into Command Prompt, Powershell, Terminal, or any other integrated terminal of your code editor.
The heart of
kel is your
config.toml holds all your personal settings, style, LLM to use, and more.
Configure your LLM's API keys in your operating system's environment variables.
Once the API key and
config.toml is configured; invoke the AI from your CLI.
kel "git command to rebase"
kel "command to get active connections in linux"
kel "What was the population of India in 1990?"
>>> I'm sorry, I can only assist with questions related to software engineering and command line tools.
I am unable to provide information on the population of India in 1990.
Now change the prompt and ask the same question.
kel "What was the population of India in 1990?" -p "You are a demography expert"
>>> The population of India in 1990 was around 874 million people.
Now change the LLM and ask the same question.
kel "What was the population of India in 1990?" -p "You are a demography expert" -c ollama -m llama2
OpenAI assistant helps in interacting with your documents to get insights, analyze, and more.
kel -a perfGPT -f <file_name>
This will enter you into chat mode with your assistant to ask questions.
Upon exiting, the assistant and files will be deleted.