Nigerian News Paraphraser & Twitter Bot

Alan Zouss

Completed work

AI Artist

BeautifulSoup

Python

Digital Media

Nigerian News Paraphraser & Twitter Bot

This project is a Python automation tool that:

Scrapes top headlines from a Nigerian news website (Punch).

Paraphrases the headline and first paragraph using a Hugging Face T5 model.

Posts the paraphrased content to X (Twitter) with relevant hashtags.

Logs all posted tweets to a CSV file for tracking and auditing.

Features

Web Scraping: Fetches the latest headlines and first paragraphs from Punch (https://punchng.com/).

AI Paraphrasing: Uses the Hugging Face model Vamsi/T5_Paraphrase_Paws to paraphrase news content and append hashtags like #NigeriaNews.

Twitter Automation: Posts paraphrased news to X (Twitter) using the official Twitter API v2 via Tweepy.

Logging: Saves each tweet's timestamp, original text, paraphrased text, and tweet URL to log.csv.

CLI Support: Specify the number of headlines to process with a command-line argument (-n).

Extensible: Modular code structure for easy adaptation to other news sites or social platforms.

Error Handling: Basic error handling and logging for robust operation.

Tech Stack

Python 3.10+

Libraries:

requests (HTTP requests)

beautifulsoup4 (HTML parsing)

tweepy (Twitter API v2)

transformers, torch, sentencepiece, tiktoken, protobuf (Hugging Face model support)

csv, logging, argparse, datetime (standard library)

No frameworks or databases required

Installation

1. Clone the repository

git clone <your-repo-url>
cd upwork

2. Install dependencies

This project uses pip for dependency management. Install all required packages with:

pip install -r requirements.txt

3. Configure API Keys

Open main.py and replace the placeholder values at the top with your actual Twitter API credentials:

TWITTER_API_KEY = "your-key"
TWITTER_API_SECRET = "your-secret"
TWITTER_ACCESS_TOKEN = "your-access-token"
TWITTER_ACCESS_SECRET = "your-access-secret"

Note: For production, consider using environment variables or a .env file for secrets.

Usage

Run the script from the project directory:

python main.py -n 3

-n or --num: Number of headlines to process (default: 3)

What happens:

The script scrapes the latest headlines and first paragraphs from Punch.

Each is paraphrased and posted to your X (Twitter) account.

All activity is logged in log.csv.

Environment Variables

Currently, API keys are hardcoded in main.py. For better security, you can refactor to use environment variables:

import os
TWITTER_API_KEY = os.environ.get('TWITTER_API_KEY')
TWITTER_API_SECRET = os.environ.get('TWITTER_API_SECRET')
TWITTER_ACCESS_TOKEN = os.environ.get('TWITTER_ACCESS_TOKEN')
TWITTER_ACCESS_SECRET = os.environ.get('TWITTER_ACCESS_SECRET')

Example .env file:

TWITTER_API_KEY=your-key
TWITTER_API_SECRET=your-secret
TWITTER_ACCESS_TOKEN=your-access-token
TWITTER_ACCESS_SECRET=your-access-secret

And load with python-dotenv.

Folder Structure

upwork/
│
├── main.py              # Main script: scraping, paraphrasing, tweeting, logging
├── requirements.txt     # Python dependencies
├── README.md            # Project documentation
├── log.csv              # (Generated) Log of posted tweets
├── pyvenv.cfg           # Python virtual environment config
├── bin/, include/, lib/ # Virtual environment folders

API Overview

No REST API is exposed.

The script interacts with:

Punch (news scraping)

Twitter API v2 (posting tweets)

Docker

No Dockerfile detected. To dockerize, create a Dockerfile with:

FROM python:3.10
WORKDIR /app
COPY . .
RUN pip install -r requirements.txt
CMD ["python", "main.py", "-n", "3"]

Contributing

Code is formatted for clarity and modularity.

No linting, formatting, or Husky hooks detected.

PRs and issues welcome!

License

MIT

Like this project

Completed work

Posted Jul 23, 2025

Automated tool for paraphrasing and tweeting Nigerian news headlines.

Likes

Views