Larry David Bot 🤖

A Bluesky and X/Twitter bot that automatically posts fictional Larry David quotes every 3 hours and 34 minutes, as if he were commenting on modern life in 2025 with his signature neurotic and comedic perspective.

Features

• Cross-platform posting: Posts to both Bluesky and Twitter (X)
• Automatic scheduling: Posts every 3 hours and 34 minutes using a scheduler
• AI-generated quotes: Uses Google's Gemini to create unique, in-character Larry David quotes
• Duplicate prevention: Caches recent posts to avoid repeats
• Modern context: Quotes reference current technology (AirPods, TikTok, AI, Zoom, etc.)
• Easy deployment: Ready to deploy,
• Fallback system: Includes backup quotes if AI generation fails

Example Quotes

• "You know what I hate? When you're at a restaurant and the server says 'Enjoy your meal' and you say 'You too'."
• "I don't trust anyone who's nice to me but rude to the waiter. Because they're just waiting until they can be rude to me too."
• "I don't like to make plans for the day because then the word 'premeditated' gets thrown around in the courtroom."
• "You know what's interesting about politics? It's not interesting."

Setup

Prerequisites

1. Bluesky Account: You need a Bluesky account and an App Password
2. Google Cloud API Key: For generating quotes with Gemini
3. Twitter API Access (Optional): For cross-posting to Twitter (X) using API v2

Local Development

1. Clone the repository:

   git clone <your-repo-url>
   cd Larry-David-Bot

2. Install dependencies:

   pip install -r requirements.txt

3. Set up environment variables:

   cp env.example .env

   Edit .env with your credentials:

   # Required
   BLUESKY_HANDLE=your-handle.bsky.social
   BLUESKY_APP_PASSWORD=your-app-password
   GEMINI_API_KEY=your-gemini-api-key
   
   # Twitter API v2 (Optional)
   TWITTER_BEARER_TOKEN=your-bearer-token
   TWITTER_API_KEY=your-api-key
   TWITTER_API_SECRET=your-api-secret
   TWITTER_ACCESS_TOKEN=your-access-token
   TWITTER_ACCESS_SECRET=your-access-secret
   

4. Run the bot:

   python larry_david_bot.py

Bluesky App Password Setup

1. Go to Bluesky Settings
2. Navigate to "App Passwords"
3. Create a new app password
4. Use this password in your .env file

Setting Up Twitter API v2

1. Go to the Twitter Developer Portal
2. Create a new Project and App
3. Under "User authentication settings":
   • Enable OAuth 2.0
   • Set App permissions to "Read and Write"
   • Set Type of App to "Web App, Automated App or Bot"
   • Set Callback URI to https://localhost
   • Set Website URL to https://github.com/Cauchon/Larry-David-Bot
4. Go to "Keys and tokens" and generate:
   • API Key and Secret
   • Access Token and Secret
   • Bearer Token (under "Authentication Tokens")

Testing Your Setup

1. Run the test script to verify your configuration:

   python test_bot.py

2. The bot will test:

   • Environment variables
   • Gemini API connection
   • Fallback quotes
   • Twitter API configuration (if provided)

Deployment on Render.com

1. Connect your repository to Render.com

2. Create a new Worker Service

3. Configure the service:

   • Build Command: pip install -r requirements.txt
   • Start Command: python larry_david_bot.py
   • Environment: Python

4. Add environment variables in Render dashboard:

   • BLUESKY_HANDLE
   • BLUESKY_APP_PASSWORD
   • GEMINI_API_KEY
   • TWITTER_BEARER_TOKEN (if using Twitter API v2)
   • TWITTER_API_KEY (if using Twitter API v2)
   • TWITTER_API_SECRET (if using Twitter API v2)
   • TWITTER_ACCESS_TOKEN (if using Twitter API v2)
   • TWITTER_ACCESS_SECRET (if using Twitter API v2)

5. Deploy! The bot will start posting every 3 hours and 34 minutes.

Configuration

Posting Schedule

The bot posts every 3 hours and 34 minutes by default. To change this, modify the schedule in larry_david_bot.py:

schedule.every(214).minutes.do(self.post_quote)

Quote Generation

• Character limit: 280 characters (Twitter-compatible)
• AI model: gemini-flash-latest
• Temperature: Default (creative)
• Fallback quotes: 5 pre-written quotes if AI fails

Duplicate Prevention

• Cache size: Last 100 posts
• Storage: JSON file (recent_posts.json)
• Retry attempts: Up to 10 attempts for unique quotes

File Structure

Larry-David-Bot/
├── larry_david_bot.py          # Main bot script
├── test_bot.py           # Test script for verification
├── requirements.txt       # Python dependencies
├── render.yaml           # Render.com deployment config
├── env.example           # Environment variables template
├── README.md            # This file
└── recent_posts.json    # Generated cache file

Logging

The bot logs all activities to:

• Console output: Real-time logs
• File: larry_david_bot.log

Customization

Adding More Fallback Quotes

Edit the get_fallback_quote() method in larry_david_bot.py:

fallback_quotes = [
    "Your new quote here, Jeff!",
    # ... existing quotes
]

Modifying the AI Prompt

Edit the generate_larry_quote() method to change the quote style or topics.

Changing Post Frequency

Modify the scheduler in run_scheduler() method.

Troubleshooting

Common Issues

1. Authentication errors: Check your Bluesky handle and app password
2. API rate limits: The bot includes retry logic for Gemini API
3. Duplicate posts: Check the recent_posts.json cache file
4. Deployment issues: Ensure all environment variables are set in Render

Logs

Check the logs for detailed error information:

tail -f larry_david_bot.log

Contributing

Feel free to submit issues or pull requests to improve the bot!

License

This project is open source. Feel free to use and modify as needed.

---

"You know what I like about this bot? It's like having me in your pocket, but without the social anxiety." - Larry David