prompt-pilot 0.1.3

Library for prompt engineering

Prompt Pilot

DISCLAIMER: This is a work in progress, and I'm sharing it as a temporary solution with some functions that you can experiment with.

Prompt Pilot is a library designed for prompt engineering in AI models. It introduces the concept of "prompt functions" to streamline the prompt creation process. With just one function, developers can create complex prompts effortlessly.

Here's an example implementation for creating a digital marketing chatbot:

# import the required functions from prompt_pilot.
from prompt_pilot.assistants import life_coach
from prompt_pilot.api_ai import gpt3
#import config function from decouple to read api_key
import decouple 

# use promptpilot to generate prompt
prompt = life_coach(name="Rocky")
print(prompt)

# configure your apikey
openai_api_key = decouple.config("OPENAI_API_KEY")

# use promptpilot's api call functions to pass prompt and apikey
output = gpt3(prompt=prompt, api_key=openai_api_key)
print(output

By automating both prompt engineering and API calls to AI models, Prompt Pilot significantly reduces the code required to build a chatbot. With just 5-6 lines of code, you can have a fully functional chatbot up and running in very little time.

Here's what you can build with prompt-pilot:

• Chatbots and other NLP tasks : Personalized chatbots for business usecase, personal usecase, and academic usecase. API calls to OpenAI and LLMs on Replicate
• Image models : Varying styles for image generators. API calls to image models on Replicate
• Data science : Automating data-science by incorporating LLMs (OpenAI) into the process to perform analysis