A powerful web application that transforms RSS feeds into clean markdown content and repurposes it for multiple platforms using AI-powered content generation.
- Real-time RSS parsing - Fetch and parse RSS feeds from any URL
- Complete article extraction - Retrieves all articles without pagination limits
- Clean markdown conversion - Converts HTML content to properly formatted markdown
- Cover image extraction - Automatically detects and displays article cover images
- Metadata preservation - Maintains author, date, and publication information
- Smart suggestions - Auto-complete with recently used RSS feeds
- Feed history - Stores up to 10 recent feeds with metadata
- Quick access - One-click loading of previously used feeds
- Feed management - Remove individual feeds or clear all history
- Text search - Search through article titles and content
- Author filtering - Filter articles by specific authors
- Date range filtering - Filter articles by publication date
- Real-time filtering - Instant results as you type
- Multiple AI providers - Choose between Groq (Llama 3.3 70B), Gemini (2.0 Flash), OpenAI (GPT-5, GPT-4o), and Anthropic (Claude 3 Opus/Sonnet)
- GPT-5 support - Latest cutting-edge model from OpenAI as default for premium content generation
- Platform-specific optimization - Tailored content for each social media and blogging platform
- Advanced persona training - Train custom AI personas with sentiment analysis, readability metrics, and semantic clustering
- Custom post types - DevRel, technical, tutorial, opinion, news, and custom styles
- Keyword integration - Natural keyword inclusion in generated content
- Professional formatting - Clean, emoji-free content focused on professional presentation
- Multi-modal preferences - Analyze and replicate content structure, formatting, and engagement patterns
- Summary generation - Concise article summaries
- Workflow diagrams - Mermaid diagram generation from content
- Image Generation - Generate images using free models (Pollinations AI), premium models (DALL-E by OpenAI), or open-source models from HuggingFace
- Author content generation - Generate content in the style of specific authors using trained personas
- Persona training - Train AI on writing samples with advanced analytics including sentiment analysis, readability metrics, and semantic clustering
- LinkedIn - Professional posts with engagement hooks (no emojis, clean formatting)
- X/Twitter - Thread-optimized content with hashtags (professional tone)
- Instagram - Clean captions with strategic hashtag placement
- Facebook - Conversational posts for community engagement
- Discord - Community-friendly messaging with markdown support
- Reddit - Platform-appropriate discussion starters
- YouTube - Video descriptions with timestamps
- TikTok - Short-form video scripts
- Medium - Article introductions and outlines
- Dev.to - Developer-focused technical content
- Hashnode - Technical blog posts with proper formatting
- Responsive design - Works seamlessly on desktop and mobile
- Dark/light mode - Automatic theme detection
- Auto-scroll - Automatically scrolls to generated content
- Copy functionality - One-click copying of markdown and generated content
- Loading states - Clear feedback during processing
- Error handling - Graceful error messages and recovery
- API key management - Secure browser-based storage for OpenAI and Anthropic keys
- Model selection - Dynamic model selection with fallback options
- Professional output - Clean, emoji-free content optimized for business use
- Next.js 15 - React framework with App Router
- TypeScript - Type-safe development
- Tailwind CSS - Utility-first styling
- Shadcn/ui - Modern component library
- Lucide React - Beautiful icons
- Next.js API Routes - Serverless API endpoints
- XML2JS - RSS feed parsing
- Turndown - HTML to Markdown conversion
- AI SDK - Unified AI provider interface
- Groq API - Ultra-fast inference with Llama models
- Google Gemini - Advanced reasoning capabilities
- OpenAI - GPT-5 (default), GPT-4o, GPT-4o-mini for premium content generation
- Anthropic - Claude 3 Opus and Sonnet for high-quality long-form content
- HuggingFace - Open-source image generation models
- Advanced persona analysis - Sentiment analysis, readability metrics, semantic clustering
- Multi-modal content support - Text, image, and formatting preference analysis
- LocalStorage - Client-side feed history persistence
- Node.js 18+
- npm or yarn package manager
Create a .env.local file in the root directory:
# Required for default AI providers
GROQ_API_KEY=your_groq_api_key_here
GEMINI_API_KEY=your_gemini_api_key_here
# Optional - for premium AI models (can also be added via UI)
OPENAI_API_KEY=your_openai_api_key_here
ANTHROPIC_API_KEY=your_anthropic_api_key_here
HUGGINGFACE_API_KEY=your_huggingface_api_key_hereNote: OpenAI and Anthropic API keys can be managed through the application UI and are stored securely in your browser's local storage.
- Clone the repository:
git clone https://github.com/your-username/rss-markdown-converter.git
cd rss-markdown-converter- Install dependencies:
npm install-
Set up environment variables (see above)
-
Run the development server:
npm run dev- Open http://localhost:3000 in your browser
npm run build
npm start- Enter RSS URL - Paste any RSS feed URL into the input field
- Browse suggestions - Use recent feeds for quick access
- Select article - Click on any article from the parsed feed
- Choose AI provider - Select between Groq, Gemini, OpenAI (GPT-5), or Anthropic
- Configure settings - Set post type, keywords, and persona options
- Generate content - Click on any platform button
- Copy and use - Copy the generated content for your platform
- GPT-5 - Latest OpenAI model for cutting-edge content generation
- GPT-4o/4o-mini - Proven OpenAI models for reliable content
- Claude 3 Opus/Sonnet - Anthropic models for nuanced, long-form content
- Gemini 2.0 Flash - Google's fast, capable model
- Llama 3.3 70B - High-performance open-source model via Groq
- Custom personas - Train AI on specific writing styles using sample content
- Advanced analytics - Sentiment analysis, readability metrics, semantic clustering
- Writing pattern analysis - Tone, structure, vocabulary, and engagement patterns
- Multi-modal preferences - Image usage, formatting styles, and content structure
- Built-in personas - Pre-trained personas for popular tech writers and DevRel professionals
- No emoji policy - All content generated without emojis for professional use
- Clean formatting - Focus on text quality and readability
- Strategic link placement - Links included only when they add value
- Platform optimization - Content tailored for each platform's best practices
- Use the search box to find specific articles
- Filter by author using the dropdown
- Set date ranges for time-specific content
- Combine filters for precise results
- Choose from predefined post types (DevRel, Technical, Tutorial, etc.)
- Add custom post types for specific needs
- Include keywords for SEO optimization
- Switch between AI providers for different writing styles
- Train custom personas with your own writing samples
- Apply advanced writing pattern analysis for consistent voice
- Recent feeds appear as suggestions while typing
- Click on any suggestion to instantly load that feed
- Remove individual feeds or clear all history
- Feed metadata shows article count and last used date
| Model | Best For | Speed | Quality | Cost |
|---|---|---|---|---|
| GPT-5 | All content types (default) | Fast | Excellent | Premium |
| GPT-4o | Blog content, complex analysis | Medium | Very High | High |
| GPT-4o-mini | Social posts, quick generation | Very Fast | High | Low |
| GPT-4-turbo | Legacy support | Medium | High | Medium |
| Model | Best For | Speed | Quality | Cost |
|---|---|---|---|---|
| Claude 3 Opus | Long-form blog content | Slow | Excellent | Premium |
| Claude 3.5 Sonnet | Balanced content generation | Medium | Very High | Medium |
| Claude 3.5 Haiku | Quick social posts | Fast | Good | Low |
| Model | Provider | Best For | Speed | Quality |
|---|---|---|---|---|
| Llama 3.3 70B | Groq | Technical content | Ultra Fast | High |
| Gemini 2.0 Flash | Creative content | Fast | High |
- Professional Blogs: GPT-5 or Claude 3 Opus
- Social Media: GPT-5 or Claude 3.5 Sonnet
- Quick Posts: GPT-4o-mini or Llama 3.3 70B
- Creative Content: Gemini 2.0 Flash or GPT-5
- Technical Writing: GPT-5 or Llama 3.3 70B
- Default model - GPT-5 is now the default for all OpenAI content generation
- Cutting-edge AI - Access to the latest language model capabilities
- Improved quality - Enhanced reasoning and content generation
- Backward compatibility - Fallback to GPT-4o and other models as needed
- Advanced analytics - Sentiment analysis, readability scoring, semantic clustering
- Writing pattern extraction - Tone, structure, vocabulary analysis
- Multi-modal support - Image preferences, formatting styles, engagement patterns
- Adaptive learning - System learns and improves from feedback
- Built-in personas - Pre-trained models for tech industry personalities
- No emojis policy - All generated content is emoji-free for professional use
- Clean formatting - Focus on text quality and readability
- Strategic link usage - Links only included when they add significant value
- Platform-specific optimization - Tailored guidelines for each social media and blogging platform
- Multiple providers - Pollinations AI (free), OpenAI DALL-E (premium), HuggingFace models
- Workflow integration - Generate images directly from article content
- Quality options - Choose between speed and quality based on your needs
- Dynamic defaults - Best model automatically selected based on content type
- Provider comparison - Easy switching between AI providers
- Custom model support - Add your own model configurations
- Performance optimization - Faster generation with smart model routing
POST /api/rss-parser
Content-Type: application/json
{
"url": "https://example.com/rss"
}
POST /api/ai-generate
Content-Type: application/json
{
"content": "Article content",
"title": "Article title",
"type": "linkedin",
"keywords": "react, javascript",
"postType": "technical",
"provider": "openai",
"model": "gpt-5",
"apiKey": "your_api_key_here",
"personaTrainingData": "optional_training_content",
"extractedLinks": [{"url": "...", "text": "..."}],
"includeSourceLink": true
}
POST /api/persona-training
Content-Type: application/json
{
"content": "Training content samples",
"personaName": "custom_persona",
"analysisType": "advanced"
}
POST /api/image-generate
Content-Type: application/json
{
"prompt": "Image description",
"provider": "openai",
"model": "dall-e-3",
"size": "1024x1024",
"quality": "hd"
}
We welcome contributions! Please follow these steps:
- Fork the repository
- Create a feature branch:
git checkout -b feature/amazing-feature - Commit your changes:
git commit -m 'Add amazing feature' - Push to the branch:
git push origin feature/amazing-feature - Open a Pull Request
- Follow TypeScript best practices
- Use Tailwind CSS for styling
- Maintain responsive design principles
- Add proper error handling
- Include loading states for async operations
This project is licensed under the MIT License - see the LICENSE file for details.
If you encounter any issues or have questions:
- Check the Issues page
- Create a new issue with detailed information
- Contact the maintainer through GitHub
- GPT-5 integration as default model
- Advanced persona training with analytics
- Professional content standards (no emojis)
- Multi-provider image generation
- Enhanced API key management
- Sentiment analysis and readability metrics
- Semantic clustering for content analysis
- Multi-modal preference learning
- Bulk content generation for multiple platforms
- Content scheduling and automation
- Advanced persona feedback system
- Real-time collaboration features
- Analytics dashboard for generated content
- Custom AI model integration beyond major providers
- Team collaboration and workspace features
- Content templates and presets library
- Export functionality (PDF, DOCX, multiple formats)
- Integration with social media APIs for direct posting
- Content performance tracking and optimization
- Multi-language support and translation
- Voice-to-content generation capabilities
Built with ❤️ by RS-labhub