Author: MrMikeAde
GitHub: github.com/MrMikeAde
Version: 2.0
License: MIT
This Python script automatically:
- Fetches articles from multiple RSS feeds (News, Tech, Business, Entertainment)
- Rewrites content using Groq's AI (Llama3-70B) to ensure originality
- Optimizes for SEO with custom prompts per feed category
- Posts drafts to your Blogger account with:
- Proper attribution
- Featured images (when available)
- SEO metadata (titles, descriptions, keywords)
| Feature | Implementation Details |
|---|---|
| Multi-Source Aggregation | Processes 5+ RSS feeds simultaneously |
| AI Content Enhancement | Uses Groq's Llama3-70B for human-like rewriting |
| Category-Specific Optimization | Custom prompts for each feed (News/Tech/Business/etc.) |
| Threaded Processing | Handles 3 feeds in parallel (configurable) |
| Smart Deduplication | Tracks processed articles to avoid reposts |
| Blogger Integration | Automatic draft creation with labels |
graph TD
A[Fetch RSS Feeds] --> B[Clean/Parse Content]
B --> C{Word Count Check}
C -->|≥15 words| D[AI Rewriting]
C -->|<15 words| E[Log as Skipped]
D --> F[SEO Enhancement]
F --> G[Blogger Draft Creation]
G --> H[Summary Reporting]
AI Rewriting Process
Input Analysis:
Extracts title, body, and source URL
Identifies feed category (for prompt selection)
Prompt Engineering:
python
# Example Tech Article Prompt:
"You are a tech journalist... rewrite focusing on innovation trends...
Include 3-5 keywords like 'AI', 'startups'... Maintain 500+ words"
Output Formatting:
Generates SEO meta description (150-160 chars)
Suggests keyword tags
Ensures natural keyword density (1-2%)
🚀 Custom Recommendations
For Content Creators
Prompt Customization:
python
# Edit CUSTOM_PROMPTS in config to match your brand voice:
"Rewrite as a [friendly/technical/sensational] writer for [your_blog_url]..."
Feed Selection:
Replace default RSS feeds with niche-specific sources
Balance categories (e.g., 3 news + 2 tech + 1 entertainment)
Post-Publishing:
Review drafts for tone consistency
Add custom featured images if auto-extracted ones are low quality
Schedule posts at optimal times (use Blogger's scheduler)
For Developers
Performance Tuning:
python
# config.py
MAX_WORKERS = 4 # Increase if you have strong CPU
ARTICLES_PER_FEED = 3 # Reduce during testing
Error Handling:
Monitor skipped_articles.txt for patterns
Implement retry logic for failed Blogger API calls
Advanced Features:
Add PDF/Word export option
Integrate with WordPress (via REST API)
Implement sentiment analysis for content filtering
⚙️ System Requirements
Minimum
yaml
Python: 3.8+
Memory: 4GB RAM
Storage: 100MB
Dependencies: See requirements.txt
API Keys: Groq + Google Blogger
Recommended
yaml
CPU: 4+ cores (for parallel processing)
Network: Stable 10Mbps+ connection
Blogger: Established blog (avoid new accounts)
Groq: Paid tier for heavy usage
📈 Performance Metrics
bash
[Typical Run]
✔ Processes 20 articles in ~8 minutes
✔ Maintains 3-5s delay between API calls
✔ Uses ~300MB memory with 3 workers
✔ Generates 500-800 word articles
⚠️ Best Practices
Legal Compliance:
Always keep drafts in "review" state
Manually verify rewritten content
Respect source websites' terms
SEO Health:
Avoid keyword stuffing
Vary article lengths (500-1200 words)
Mix AI content with human-written posts
Maintenance:
Rotate RSS sources monthly
Update prompts quarterly
Clear skipped_articles.txt weekly
🛠️ Setup Guide
Install dependencies:
bash
pip install -r requirements.txt
Configure .env:
ini
GROQ_API_KEY=your_key_here
BLOGGER_BLOG_ID=123456789
BLOGGER_BLOG_URL=yourblog.blogspot.com
Run:
bash
python rss_to_blogger.py