🔥 DropScore

Reverse-engineer virality in seconds.

Paste a YouTube link — DropScore scrapes the top comments, runs sentiment and keyword analysis, and reveals what made the video blow up. Great for creators, editors, marketers, or just curious minds.

I kept asking a content strategist how he makes videos go viral.
He always said: "It’s just song choice."

That didn’t sit right. So I spent a weekend building this to find out for myself.

Turns out the truth lies in the comments — raw emotional signals, hype moments, reactions, inside jokes. DropScore decodes all that using lightweight NLP, clean visualizations, and a bit of Gen-Z flair.

• ✅ Scrape top YouTube comments via YouTube Data API
• ✅ Sentiment scoring using TextBlob
• ✅ Keyword clustering with spaCy
• ✅ AI-style local summary (free, no OpenAI key needed)
• ✅ Viral factor score based on comment volume, mood, and variance
• ✅ Meme Mode 🧠 (adds falling emoji and Comic Sans — because why not?)
• ✅ Built with Streamlit — runs locally, looks sleek

git clone https://github.com/yourusername/DropScore.git
cd DropScore

python3 -m venv venv
source venv/bin/activate  # On Windows: venv\Scripts\activate

pip install -r requirements.txt
python -m spacy download en_core_web_sm

Create a .env file in the root folder with this:

YOUTUBE_API_KEY=your_api_key_here

You can get a free API key from: https://console.cloud.google.com/apis/library/youtube.googleapis.com

streamlit run app.py

Open http://localhost:8501 in your browser.

• 🔍 Scraper: Pulls 100 top-level YouTube comments using official API
• 📊 Analyzer: Uses TextBlob for sentiment, spaCy for keyword lemmatization, and Counter for frequency
• 🧠 Summary Engine: Synthesizes a quick AI-style writeup with Gen-Z flair
• 📈 Viral Factor: A custom score based on mood, volume, and engagement variation
• 🎭 Meme Mode: Optional toggle with emojis falling from the sky and Comic Sans aesthetics

• Re-enable TikTok support with a more stable API
• Parse thumbnails and titles for visual cues
• Smart topic detection via LLM or BERTopic
• Downloadable PDF reports for creators
• Export-to-Notion / Copy-to-Clipboard for summaries

• Streamlit
• TextBlob
• spaCy
• Google YouTube Data API
• Comic Sans and emoji rain for the vibes

Built by David Lai
🎻 Former concertmaster · 🧠 Start-Up strategist

DropScore: Don’t predict the next trend. Decode the last one.