Paper Summarizer Slack Bot

A Slack bot that summarizes academic papers using GPT-3o, following the methodology outlined in "How to read a paper" by S. Keshav.

Features

• Triggered by /summary command in a Slack thread
• Extracts paper links from the parent message
• Generates a comprehensive summary using GPT-3o
• Follows Keshav's paper reading methodology:
  • First pass: 5C details (Category, Context, Correctness, Contributions, Clarity)
  • Second pass: Detailed analysis of figures, methods, and key points

Project Structure

paper-summarizer/
├── docs/                       # Documentation
│   └── slack_app_setup.md      # Guide for setting up Slack app
├── scripts/                    # Utility scripts
│   └── deploy.sh               # Deployment script
├── src/                        # Source code
│   ├── __init__.py             # Package initialization
│   ├── app.py                  # Main Flask application
│   ├── paper_processor.py      # Paper extraction and processing
│   ├── slack_client.py         # Slack API interactions
│   ├── summarizer.py           # GPT-3o integration for summaries
│   └── test_summarizer_locally.py # Local testing script
├── tests/                      # Test suite
│   ├── __init__.py             # Test package initialization
│   └── test_paper_processor.py # Tests for paper processor
├── .dockerignore               # Docker ignore file
├── .env.example                # Example environment variables
├── .gitignore                  # Git ignore file
├── Dockerfile                  # Docker configuration
├── docker-compose.yml          # Docker Compose configuration
├── Makefile                    # Makefile for common commands
├── README.md                   # This file
├── requirements.txt            # Project dependencies
├── run_tests.py                # Script to run tests
└── setup.py                    # Package setup script

Setup

Prerequisites

• Python 3.8 or higher
• A Slack workspace with admin privileges
• An OpenAI API key with access to GPT-3o

Installation

1. Clone this repository:

   git clone https://github.com/yourusername/paper-summarizer.git
   cd paper-summarizer

2. Create and activate a virtual environment:

   make venv
   source venv/bin/activate  # On Windows: venv\Scripts\activate

3. Install dependencies:

   make setup

4. Create a .env file from the example:

   cp .env.example .env

5. Edit the .env file with your credentials:

   SLACK_BOT_TOKEN=xoxb-your-slack-bot-token
   SLACK_SIGNING_SECRET=your-slack-signing-secret
   OPENAI_API_KEY=your-openai-api-key
   PORT=3000
   

6. Set up your Slack app following the instructions in docs/slack_app_setup.md

Running the Application

Local Development

Start the application:

make run

Or manually:

python src/app.py

Docker Deployment

You can also run the application using Docker:

# Build and start the container
docker-compose up -d

# View logs
docker-compose logs -f

# Stop the container
docker-compose down

Make sure your .env file is set up with the required environment variables before running with Docker.

Server Deployment

A deployment script is provided to help you deploy the application to a remote server:

1. Edit the configuration in scripts/deploy.sh to match your server details:

   REMOTE_USER="your-username"
   REMOTE_HOST="your-server.com"
   REMOTE_DIR="/path/to/deployment"

2. Run the deployment script:

   ./scripts/deploy.sh

The script will build the Docker image locally, copy the necessary files to your remote server, and start the application using Docker Compose.

Testing the Summarizer Locally

You can test the summarizer without setting up a Slack app:

# Using the Makefile
make test-summarizer URL=https://arxiv.org/abs/1706.03762

# Or directly
python src/test_summarizer_locally.py https://arxiv.org/abs/1706.03762

This will process the paper at the given URL and print the summary to the console.

Development

Run tests:

make test

Clean up build artifacts:

make clean

Format code with Black (if installed):

make format

Run linting (if flake8 is installed):

make lint

Run tests with coverage (if pytest-cov is installed):

make coverage

Usage

1. Share an academic paper link in a Slack channel
2. In a thread on that message, type /summary
3. The bot will analyze the paper and post a summary in the thread

Summary Format

The summary follows Keshav's paper reading methodology:

First Pass: The Five Cs

• Category: Type of paper (measurement, analysis, research prototype, etc.)
• Context: Related papers and theoretical bases
• Correctness: Validity of assumptions
• Contributions: Main contributions of the paper
• Clarity: Writing quality assessment

Second Pass: Detailed Analysis

• Analysis of figures, diagrams, and illustrations
• Evaluation of methodology and results
• Comprehensive summary of key findings and implications

License

MIT

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.