4 steps to create chat bot for your API

Description how to create LLM chat bit for existing API using Langchain, Streamlit and Python.

Start

1. Virtual Environment

pipenv shell pipenv install

1. Update .env file Use your OpenAI token https://help.openai.com/en/articles/4936850-where-do-i-find-my-openai-api-key

2. Start Chat application

pipenv run streamlit run ./04-chatbot.py

1. Start chating

Please list all names of pets in the store with status sold.

Packages

Main modules:

• langchain
• tiktoken
• langchain-community

Example Petshop API:

Swagger UI: https://petstore.swagger.io/#/

!wget https://petstore.swagger.io/v2/swagger.yaml -O petshop_openapi.yaml

Readings

• (LangChaing)[https://www.langchain.com/]
• (OpenAPI Agent)[https://python.langchain.com/docs/integrations/toolkits/openapi]
• (Streamlit)[https://docs.streamlit.io/]
• (Pet Shop API)[https://petstore.swagger.io/#/]