README.md

metadata

title: MSDSF25M005_Ver2— Website & YouTube Q&A
emoji: 🤖
colorFrom: blue
colorTo: green
sdk: gradio
sdk_version: 4.44.0
app_file: app.py
pinned: false

Version 2 — Website Scraper & YouTube Transcript Q&A

A Gradio app with two independent input modes:

Tab 1: Bot-Protected Website Scraper

• Enter any URL in the text box
• Fetches and parses the page using Bright Data Web Unlocker API + BeautifulSoup
• Extracted content is used as context for Q&A with a Groq-hosted LLM

Tab 2: YouTube Transcript Q&A

• Enter a valid YouTube Video ID (e.g., dQw4w9WgXcQ)
• Uses youtube-transcript-api to fetch auto-generated or manual transcripts
• Transcript is passed to Groq LLM for Q&A (e.g., "What is the main topic?", "Summarize in 3 bullet points")
• Displays a clear error if no transcript is available

Setup (Hugging Face Spaces)

Add these secrets in your Space Settings → Repository secrets:

• GROQ_API_KEY — Your Groq API key
• BRIGHTDATA_API_KEY — Your Bright Data Web Unlocker API key (for Tab 1)

Create Hugging Face Space

1. Go to huggingface.co/new-space
2. Choose SDK: Gradio, Space name: scraper_bot_v2 (or any name)
3. Clone the Space, then copy app.py, requirements.txt, and README.md from this folder into it
4. In Space Settings → Repository secrets, add:
   • GROQ_API_KEY
   • BRIGHTDATA_API_KEY
5. Push to the Space — it will build and deploy automatically

Local Run

pip install -r requirements.txt

Create a .env file with:

GROQ_API_KEY=your_groq_key
BRIGHTDATA_API_KEY=your_brightdata_key

Then:

python app.py