A real-time, AI-powered leaderboard of GitHub repository activity with live play-by-play commentary! Watch as repositories compete for the top spots with an animated sportscaster announcing each event.

• Real-Time Monitoring: Monitors the latest GitHub public events globally
• Configurable Scope: Watch all of GitHub, specific organizations, repositories, or users
• Event Filtering: Filter by event types (Push, PR, Issues, Stars, Forks, Releases)

• Activity Scoring: Repositories ranked by weighted activity scores
• Trending Indicators: See which repos are climbing or falling in the rankings
• Real-Time Stats: Track total events, contributors, and activity scores
• Animated Transitions: Beautiful animations when repos change positions

• Sports-Style Announcements: AI-generated play-by-play commentary for each event
• Configurable AI Backend: Connect to OpenAI or any compatible API
• Fallback Templates: Built-in engaging commentary when AI is unavailable
• Multiple Models: Support for gpt-4o-mini, gpt-4o, o3-mini, and gpt-3.5-turbo

• Sound Effects: Unique audio cues for each event type (Push, PR, Star, Fork, etc.)
• Text-to-Speech: Have the AI commentary read aloud by the browser
• Mute Controls: Toggle sound effects and TTS independently

• All Activity: See everything happening in real-time
• 🔥 Hot: High-impact events (Pushes, PRs, Releases)
• 💻 Code: Code-related activity (Pushes, PRs, Creates, Deletes)
• 👥 Social: Community engagement (Stars, Forks, Issues, Comments)

• Animated Sportscaster: SVG robot that "announces" each event
• TV Screen Interface: Retro-inspired broadcast display
• Glassmorphic Cards: Modern, beautiful UI components
• Dark Theme: Easy on the eyes for 24/7 streaming
• Responsive Design: Works on desktop, tablet, and mobile

• Monitoring Scope: Global, Organization, Repository, or User
• Preset Channels: Quick access to popular tech categories (Web, AI/ML, DevOps)
• Refresh Speed: Adjustable from 5-60 seconds
• Auto-Protect: Automatic rate limit throttling
• Local Storage: Settings persist across sessions

Visit the live page at: https://owasp-blt.github.io/Github_Sportscaster/

The page fetches data from the GitHub Events API and processes it in real-time:

1. Event Collection: Polls GitHub API for public events
2. Smart Processing: Filters duplicates, categorizes events
3. Activity Scoring: Weights events by type and recency
4. Leaderboard Updates: Ranks repositories by activity score
5. AI Commentary: Generates sports-style announcements
6. Live Rendering: Animates changes in real-time

• Single Page App: HTML + CSS + JavaScript
• No Build Required: Works directly from static hosting
• Modular Design: Separate classes for SoundEffects, TTS, AI Commentary, and Main Sportscaster

• index.html - Main UI structure and styling
• sportscaster.js - Core JavaScript functionality

• GitHub Events API: Real-time public events
• Conditional Requests: Uses ETags to minimize API calls
• Rate Limit Protection: Auto-throttling when limits are low
• Demo Mode: Fallback to simulated data if API is unavailable

Configure your own AI API for commentary:

1. Click the ⚙️ Settings button
2. Enter your API URL (e.g., https://api.openai.com/v1/chat/completions)
3. Enter your API key
4. Select your preferred model
5. Enable AI Commentary with the 🤖 button

Simply open the page and watch! The sportscaster will:

• Update every 10 seconds (configurable)
• Show new repositories as they appear
• Highlight repos with new activity
• Announce events with AI commentary (when enabled)

1. Open Settings (⚙️)
2. Set Scope to "Organization"
3. Enter org name (e.g., "facebook")
4. Click "Apply Configuration"

1. Open Settings (⚙️)
2. Set Scope to "Repository"
3. Enter repo (e.g., "facebook/react")
4. Click "Apply Configuration"

1. Open Settings (⚙️)
2. Enter your OpenAI API URL
3. Enter your API key
4. Select a model
5. Click "Apply Configuration"
6. Toggle the 🤖 button to enable

To run locally:

# Clone the repository
git clone https://github.com/OWASP-BLT/Github_Sportscaster.git
cd Github_Sportscaster

# Serve the files (any HTTP server works)
python3 -m http.server 8080

# Open http://localhost:8080 in your browser

• API keys are stored in browser localStorage
• No server-side processing of credentials
• All API calls made directly from browser
• Consider using environment-restricted API keys

This project is open source and available under the MIT License.