vedanshagrawal2000-debug/Spam-Detector — reverse-engineered prompt

Build me a simple spam message detector web app in Python. I want a clean Streamlit page where someone can paste or type an SMS message, click a button, and instantly see whether it looks like spam or not spam.

Use the included spam.csv data to train the model. Clean the text, turn messages into numbers with a basic bag of words approach, and train a Naive Bayes classifier with scikit learn. Keep the code easy to understand and split it into simple files for the app, preprocessing, and model training. Show a friendly result on the page, with a clear warning style for spam and a safe style for normal messages.

Please include the requirements file and make sure it can run locally with pip install and streamlit run app.py. If anything has changed in the current Streamlit or scikit learn docs, look it up and use the current best way.