Twitter Top Tweets Analyzer

A web app that finds the top 10 performing tweets for any Twitter/X user based on engagement metrics.

Features

• Analyze any public Twitter/X account
• Filter by time period (last 1 or 3 months)
• Engagement scoring formula:
  • Score = Likes + (Retweets × 2) + Replies + Quotes + Bookmarks
• Ranking combines absolute engagement (70%) and engagement rate (30%)
• Clean, responsive UI

Setup

1. Get Twitter API Access

1. Go to Twitter Developer Portal
2. Create a new project and app
3. Generate a Bearer Token (under Keys and Tokens)

2. Install Dependencies

cd twitter-analyzer
python3 -m venv venv
source venv/bin/activate  # On Windows: venv\Scripts\activate
pip install -r requirements.txt

3. Configure Environment

cp .env.example .env
# Edit .env and add your TWITTER_BEARER_TOKEN

4. Run the App

uvicorn app.main:app --reload

Visit http://localhost:8000

API Endpoints

POST /api/analyze

Analyze a user's tweets programmatically.

Request:

{
  "username": "elonmusk",
  "months": 1,
  "top_n": 10
}

Response:

{
  "user": {
    "id": "...",
    "username": "elonmusk",
    "name": "Elon Musk",
    "followers_count": 100000000,
    ...
  },
  "tweets": [
    {
      "tweet_id": "...",
      "text": "...",
      "likes": 50000,
      "retweets": 10000,
      "engagement_score": 70000,
      "engagement_rate": 0.07,
      "url": "https://x.com/i/status/..."
    },
    ...
  ],
  "total_tweets_analyzed": 10,
  "time_period_months": 1
}

Limitations

• Rate Limits: Twitter API has rate limits. The Basic tier allows ~15 requests per 15 minutes for user timeline
• No Impressions: Twitter doesn't expose impression data for other users' tweets, so we use engagement metrics as a proxy
• Public Tweets Only: Can only analyze public accounts

Tech Stack

• Backend: FastAPI + Python
• Twitter API: Tweepy
• Frontend: Jinja2 templates + Tailwind CSS