🤖 YUV.AI Developers AI Trends

Stay ahead of the curve with curated Gen AI & Machine Learning trends, delivered in a beautiful digest with AI-powered summaries.

A personalized news aggregator that fetches, ranks, and presents the latest trending content from the AI/ML ecosystem in a stunning, Apple Newsroom-inspired layout. Now featuring AI-generated summaries and trending explanations powered by Cohere or Anthropic Claude.

🏗️ System Architecture

graph TB
    subgraph "User Interface"
        CLI[CLI main.py]
        HTML[Generated HTML Digest]
    end
    
    subgraph "Core Processing"
        CLI --> Fetchers[Fetchers Module]
        Fetchers --> Ranker[Ranker Module]
        Ranker --> Generator[Generator Module]
        Generator --> Summarizer[AI Summarizer]
        Summarizer --> Generator
        Generator --> HTML
    end
    
    subgraph "Data Sources"
        GitHub[GitHub Trending API]
        GitHubExplore[GitHub Explore]
        HFPapers[Hugging Face Papers]
        HFSpaces[Hugging Face Spaces API]
        
        Fetchers -->|Web Scraping| GitHub
        Fetchers -->|Web Scraping| GitHubExplore
        Fetchers -->|Web Scraping| HFPapers
        Fetchers -->|API Call| HFSpaces
    end
    
    subgraph "AI Services"
        Cohere[Cohere API<br/>command-a-03-2025]
        Anthropic[Anthropic Claude<br/>claude-sonnet-4]
        
        Summarizer -->|Generate Summaries| Cohere
        Summarizer -->|Generate Summaries| Anthropic
    end
    
    subgraph "Data Flow"
        GitHub -->|Stars, Velocity,<br/>Topics, Forks| Fetchers
        GitHubExplore -->|Collections| Fetchers
        HFPapers -->|Papers, Authors,<br/>Upvotes, arXiv IDs| Fetchers
        HFSpaces -->|Spaces, Likes,<br/>SDKs| Fetchers
        
        Fetchers -->|Raw Items| Ranker
        Ranker -->|Scored & Ranked<br/>Top N Items| Generator
        Generator -->|Items + Context| Summarizer
        Summarizer -->|AI Summary<br/>Trending Reason| Generator
        Generator -->|Jinja2 Template| HTML
    end
    
    subgraph "Configuration"
        Config[config.py]
        Config -.->|Settings| Fetchers
        Config -.->|Weights| Ranker
        Config -.->|API Keys| Summarizer
    end
    
    style CLI fill:#667eea,stroke:#333,stroke-width:2px,color:#fff
    style HTML fill:#34c759,stroke:#333,stroke-width:2px,color:#fff
    style Cohere fill:#ff9800,stroke:#333,stroke-width:2px,color:#fff
    style Anthropic fill:#764ba2,stroke:#333,stroke-width:2px,color:#fff
    style Summarizer fill:#667eea,stroke:#333,stroke-width:2px,color:#fff

Data Flow Steps

CLI Invocation → User runs python main.py with options
Fetchers → Parallel collection from multiple sources:
- GitHub Trending (web scraping)
- Hugging Face Papers (web scraping)
- Hugging Face Spaces (official API)
- GitHub Explore Collections (web scraping)
Ranker → Scores items based on:
- Popularity metrics (40%)
- Growth velocity (30%)
- Recency (30%)
Generator → Takes top N ranked items
AI Summarizer → For each item:
- Builds context from metadata
- Calls Cohere or Anthropic API
- Generates concise summary
- Explains trending reasons
Template Rendering → Jinja2 renders HTML with enriched data
Output → Beautiful HTML digest with AI summaries & copy buttons

✨ Features

📊 Multi-Source Aggregation

🔥 GitHub Trending - Hottest repos with star velocity tracking
📄 Hugging Face Papers - Latest research from arXiv & HF Daily Papers
🚀 Hugging Face Spaces - Trending interactive ML demos
📦 GitHub Explore - Curated collections (coming soon)

🎨 Beautiful Design

Clean, modern card-based layout inspired by Apple Newsroom
Responsive design that works on all devices
Smart typography and spacing
Smooth animations and hover effects

🎯 Smart Ranking

Popularity metrics (stars, likes, upvotes)
Growth velocity (stars/day)
Real-time trending indicators
Topic categorization and tags

🤖 AI-Powered Insights ✨ NEW!

Automatic one-sentence summaries for each item
AI-generated trending explanations
Powered by Cohere or Anthropic Claude
Grounded in real metrics and descriptions
One-click copy to clipboard

⚙️ Flexible & Customizable

Multiple time ranges: daily, weekly, monthly, or custom
Configurable sources and filters
Language-specific trending (Python, TypeScript, Jupyter, etc.)
Easy branding customization
Optional AI summaries (can be disabled)

🤖 Automation Ready

CLI interface for easy scripting
Supports Windows Task Scheduler, cron, GitHub Actions
One-command digest generation
Optional auto-open in browser

🚀 Quick Start

Prerequisites

Python 3.8 or higher
pip package manager
(Optional) Cohere or Anthropic API key for AI summaries

Installation

Clone the repository

git clone https://github.com/hoodini/yuv-ai-trends.git
cd yuv-ai-trends

Install dependencies

pip install -r requirements.txt

(Optional) Set up AI API key

# For Cohere (recommended - faster)
export COHERE_API_KEY="your-api-key-here"

# OR for Anthropic Claude
export ANTHROPIC_API_KEY="your-api-key-here"

# Get your free API key:
# Cohere: https://dashboard.cohere.com/api-keys
# Anthropic: https://console.anthropic.com/

Generate your first digest

# With AI summaries (if API key is set)
python main.py --range daily --open

# Without AI summaries
python main.py --range daily --open --no-ai

That's it! Your browser will open with today's AI trends digest. 🎉

📖 Usage

Basic Commands

# Daily digest (default)
python main.py --range daily

# Weekly digest
python main.py --range weekly

# Monthly digest
python main.py --range monthly

# Custom date range (14 days)
python main.py --days 14

# Limit number of items
python main.py --range daily --limit 30

# Auto-open in browser
python main.py --range daily --open

# Custom output filename
python main.py --range daily --output my_digest.html

Command-Line Options

Option	Description	Example
`--range`	Time range: daily, weekly, monthly	`--range weekly`
`--days`	Custom number of days	`--days 7`
`--limit`	Max items in digest	`--limit 50`
`--open`	Open in browser after generation	`--open`
`--output`	Custom output filename	`--output weekly.html`
`--no-ai`	Disable AI-powered summaries	`--no-ai`

⚙️ Configuration

Edit config.py to customize your digest:

# GitHub settings
GITHUB_LANGUAGES = ["python", "jupyter-notebook", "typescript"]
GITHUB_TOPICS = ["machine-learning", "deep-learning", "llm", "generative-ai"]

# Hugging Face settings
HF_SPACES_TRENDING_LIMIT = 20

# Scoring weights (adjust to your preference)
SCORING_WEIGHTS = {
    "stars_weight": 0.4,      # GitHub stars importance
    "recency_weight": 0.3,    # How recent matters
    "velocity_weight": 0.3,   # Growth rate importance
}

# AI Summarization settings
AI_SUMMARIES_ENABLED = True  # Set to False to disable globally
AI_MAX_WORKERS = 3  # Parallel API calls (lower for rate limits)

API Key Setup

Create a .env file or set environment variables:

# Option 1: Cohere (recommended - faster, generous free tier)
COHERE_API_KEY=your_cohere_key_here

# Option 2: Anthropic Claude (more detailed summaries)
ANTHROPIC_API_KEY=your_anthropic_key_here

The system will auto-detect which API key is available and use it.

🤖 Automation

Windows (Task Scheduler)

PowerShell command:

$action = New-ScheduledTaskAction -Execute "python" -Argument "C:\path\to\news\main.py --range daily --open"
$trigger = New-ScheduledTaskTrigger -Daily -At 8am
Register-ScheduledTask -Action $action -Trigger $trigger -TaskName "AITrendsDaily"

Linux/Mac (Cron)

# Edit crontab
crontab -e

# Add this line for daily 8am execution
0 8 * * * cd /path/to/news && python main.py --range daily

GitHub Actions

Create .github/workflows/daily-digest.yml:

name: Daily AI Trends
on:
  schedule:
    - cron: '0 8 * * *'  # 8am UTC daily
jobs:
  generate:
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v3
      - uses: actions/setup-python@v4
        with:
          python-version: '3.10'
      - run: pip install -r requirements.txt
      - run: python main.py --range daily

See AUTOMATION.md for more detailed automation options.

📁 Project Structure

yuv-ai-trends/
├── main.py              # CLI entry point & orchestration
├── config.py            # Configuration settings
├── fetchers.py          # Data collection from sources
├── ranker.py            # Scoring & ranking algorithms
├── generator.py         # HTML digest generation
├── summarizer.py        # 🆕 AI-powered summaries (Cohere/Anthropic)
├── hf_mcp.py           # Hugging Face MCP integration
├── templates/
│   └── digest.html      # Beautiful HTML template w/ AI sections
├── output/              # Generated digests
├── requirements.txt     # Python dependencies
├── .env.example        # API key template
├── README.md           # You are here!
└── AUTOMATION.md       # Automation guides

🎨 Customization

Branding

Edit templates/digest.html header section to customize branding:

<header>
    <div>
        <strong>Your Name</strong> • Your Title • <a href="https://yoursite.com">YourSite</a>
    </div>
    <h1>🤖 Your AI Trends Title</h1>
</header>

Styling

The template uses clean, modern CSS. Key classes to customize:

.item - Card styling
.item-title - Title appearance
.section-title - Section headers
header - Top banner gradient

🔧 Technical Details

Data Sources

GitHub Trending: Web scraping with BeautifulSoup
Hugging Face Papers: HF Papers page scraping + arXiv metadata
Hugging Face Spaces: Official HF Hub API
GitHub Explore: Web scraping (limited availability)

Ranking Algorithm

Items are scored based on:

Popularity: Stars, likes, upvotes (40%)
Velocity: Recent growth rate (30%)
Recency: How recent the item is (30%)

Scores are normalized to 0-100 and items are sorted accordingly.

Technologies

Python 3.8+: Core language
Requests: HTTP client
BeautifulSoup4: HTML parsing
Hugging Face Hub: Official API client
Jinja2: HTML templating
Click: CLI interface
Cohere SDK: AI summary generation (primary)
Anthropic SDK: Alternative AI provider
Concurrent Futures: Parallel API calls

🤝 Contributing

Contributions are welcome! Here are some ways you can help:

🐛 Report bugs and issues
💡 Suggest new features or data sources
🎨 Improve the design/layout
📖 Improve documentation
🔧 Submit pull requests

Development Setup

# Fork and clone the repo
git clone https://github.com/yourusername/yuv-ai-trends.git

# Create a virtual environment
python -m venv .venv
source .venv/bin/activate  # or .venv\Scripts\activate on Windows

# Install dependencies
pip install -r requirements.txt

# Make your changes and test
python main.py --range daily --open

📝 License

MIT License - feel free to use this project for personal or commercial purposes.

👤 Author

Yuval Avidani

🌐 Website: YUV.AI
💼 AI Builder & Speaker
🚀 Passionate about democratizing AI knowledge

🙏 Acknowledgments

GitHub for trending data
Hugging Face for Papers and Spaces API
The amazing open-source AI/ML community

📊 Sample Output

The digest includes:

Total Items: Count of all trending content
GitHub Repos: With stars, velocity, topics, forks, contributors
Papers: With authors, arXiv IDs, upvotes, publication dates
Spaces: With likes, SDKs, creation dates
✨ AI Summaries: One-sentence description of what each item does
🔥 Trending Explanations: Why it's trending and what's innovative
📋 Copy Buttons: One-click copy summary to clipboard

All organized in a beautiful, clickable layout with smooth scrolling and responsive design.

AI Summary Example

Project: UniVA: Universal Video Agent
✨ AI Summary: UniVA is an open-source, generalist AI agent designed to perform diverse video-related tasks universally.
🔥 Why Trending: UniVA is trending due to its groundbreaking ability to handle a wide range of video tasks—from editing and summarization to generation—with a single model, democratizing advanced video AI through open-source accessibility.

🔮 Roadmap

AI-powered summaries (✅ Completed!)
Multiple AI provider support (✅ Cohere + Anthropic)
Copy to clipboard functionality (✅ Completed!)
Add more data sources (Papers with Code, Reddit, Twitter)
Email digest delivery
RSS feed generation
User accounts and preferences
Mobile app
Sentiment analysis and topic clustering

Made with ❤️ for the AI/ML community

Name		Name	Last commit message	Last commit date
Latest commit History 5 Commits
output		output
templates		templates
.env.example		.env.example
.gitignore		.gitignore
AGENT_PROMPT.md		AGENT_PROMPT.md
AUTOMATION.md		AUTOMATION.md
LICENSE		LICENSE
README.md		README.md
config.py		config.py
fetchers.py		fetchers.py
generator.py		generator.py
hf_mcp.py		hf_mcp.py
main.py		main.py
ranker.py		ranker.py
requirements.txt		requirements.txt
summarizer.py		summarizer.py

License

hoodini/yuv-ai-trends

Folders and files

Latest commit

History

Repository files navigation