Skip to content

aza1200/GitHub-QA-Chatbot-with-Langchain

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

87 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Git-ter : GitHub QA Chatbot with Langchain

Git-ter is short for "Feather: Rob a Github", and has the same pronunciation as "Feather" in Korean.
Git-ter is an analytical chatbot development framework dedicated to GitHub.
Git-ter's goal is to increase the productivity and efficiency of development by helping shorten the time and effort to study through chatbot applications.

Why Git-ter

  • ChatGPT cannot learn the latest framework without web access extensions. So we created Git-ter, an application that can complement that.
  • Using Langchain, which deals with Large Language Model(LLM). You can ask and answer specialized questions about the repository you entered.
  • Store and retrieve data at high speed. Use FAISS, which is a Langchain vector store.

Target Group

Anyone who is interested in programming looks at the GitHub repository.
However, if there is too much content in the Repository, or if there is not enough explanation about the code, it's difficult to understand what it is. Or if you're new to coding, you're at a loss where to start looking at which files.

Features

Git-ter is not just a chit-chat chatbot, provides file structure, code analysis, and summarization within the repository to make the content of the GitHub repository of interest easier to learn.


DEMO Video

GITTER_DEMO


Git-ter Tutorial

Use GPT 3.5-turbo-16k for

GitHub RestAPI, LangChain, FAISS(Vector DB) , Openai, and Streamlit.

Development

  1. Clone the repo or download the ZIP
git clone [github https url]
  1. Install virtual environment and packages

This repository runs for Python==3.9.13

1. conda create -n ENV_NAME python=3.9.13
2. conda activate ENV_NAME   
3. pip install -r requirements.txt   
  1. Set up your .env file
  • make ".streamlit/secrets.toml" file
GITHUB_NAME = 
GITHUB_TOKEN = 
OPENAI_API_KEY = 
  • Visit openai to retrieve API keys and insert into your .streamlit/secrets.toml file.
  • Visit github to create github access token

Please Note

✨ Our application recommends using Chrome.
If you run into errors, Open an issue.
✨ When you install and test the repository, you should change all of the names and tokens to GITHUB_NAME and GITHUB_TOKEN at get_info_from_api.py.

# Example : get_info_from_api.py
# BEFORE
response = requests.get(url,auth=(st.secrets["GITHUB_NAME_1"], st.secrets["GITHUB_TOKEN_1"]))

# AFTER
response = requests.get(url,auth=(st.secrets["GITHUB_NAME"], st.secrets["GITHUB_TOKEN"]))

Contributers

wonyoung jaehyeong sangmin sangHui

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Python 100.0%