Skip to main content

Intro

Let's discover Kel in less than 5 minutes.

Demo​

Kel-Demo

Getting Started​

Kel is your AI assitant in your CLI.

What you'll need​

  • Python3.6 or above
  • pip
  • respective LLM's API keys

How to get started?​

Clone kel repo to your local.

cd kel
pip install .

You can type this command into Command Prompt, Powershell, Terminal, or any other integrated terminal of your code editor.

Configuration​

The heart of kel is your config.toml.

config.toml holds all your personal settings, style, LLM to use, and more.

API Keys​

Configure your LLM's API keys in your operating system's environment variables.

export OPENAI_API_KEY="sk-...AbyZ"
export ANTHROPIC_API_KEY="sk-ant-...ZZZ"

Usage​

Once the API key and config.toml is configured; invoke the AI from your CLI.

kel -v
kel -h
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​

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.