normandmickey / MrsStax

QA Slack Bot - MrsStax

Introduction

This application is a Slack Bot that uses Langchain and OpenAI's GPT3 language model to provide domain specific answers. You provide the documents.

This app includes the OpenStax textbook "Principles of Macroeconomics 3e" for demonstration purposes.

Principles of Macroeconomics 3e is licensed under a Creative Commons Attribution 4.0 International (CC BY) license, which means that you can distribute, remix, and build upon the content, as long as you provide attribution to OpenStax and its content contributors.

Features

• GPT3 based Slack Bot using your own documents (PDF or docx) for added context and increased accuracy.
• Utilizes a local vector database (FAISS-CPU)

Usage

To use the MrsStax Slack Bot, the following environment variables need to be set in your .env file:

• SLACK_BOT_TOKEN: Token for the Slack Bot.
• SLACK_APP_TOKEN: Token for the Slack app.
• OPENAI_API_TOKEN: Token for OpenAi

Installation

Requires Python3.10 or higher

Clone this repo and run the following commands

python3 -m venv venv
source venv/bin/activate
pip install -r requirements.txt
cp env.example .env

1. Set your OPENAI_API_TOKEN in the .env file.

2. Create new Slack App - https://api.slack.com

3. Click on "Basic Information"

   • Click on "Generate Token and Scopes"

     • Token Name = "App Token"
     • App Scope = "connections:write"

   • Copy "App Token" and paste it into your .env file as "SLACK_APP_TOKEN".

4. Click on "Socket Mode"

   • Click on "Enable"

5. Click on "OAuth & Permissions" and add the following permissions.

   • app_mentions:read

   • chat:write

   • chat:write.public

   • im:history

   • Copy "Bot User OAuth Token" and paste it into your .env file as "SLACK_BOT_TOKEN".

6. Click on "App Home" and make sure "Messages Tab" is enabled and check the box for "Allow users to send Slash commands and messages from the messages tab".

7. Click on "Event Subscriptions" and enable it. Click on "Subscribe to bot events" and add "message.im"

8. Install App into your Slack.

9. Upload or copy your .pdf or .docx files to the "docs" folder.

10. Run the following commands.

python ingest.py
python app.py

11. Visit your Slack and send direct message to your bot.

If you left the two original files (Macroeconomics3e-WEB.pdf, pdf1.txt) in your docs directory you should be able to ask your bot economics related questions. For example.

• Q1. Residents of the town of Smithfield like to consume hams, but each ham requires 10 people to produce it and takes a month. If the town has a total of 100 people, what is the maximum amount of ham the residents can consume in a month?

• Q2. Why might Belgium, France, Italy, and Sweden have a higher export to GDP ratio than the United States?

• Q3. What is dumping? Why does prohibiting it often work better in theory than in practice?

12. Your vector database needs to be re-indexed each time you add or remove documents from your docs folder. To do this simply run python ingest.py.