Welcome to the ragleaf documentation. Find everything you need to integrate and use our RAG quality helper.

Quick Links

Core Concepts

Understanding the fundamentals of RAG quality checking:

Citation Verification

Learn how ragleaf verifies that generated responses are properly grounded in source documents. Our citation verification system checks for:

• Factual accuracy against source material
• Proper attribution of information
• Detection of hallucinated content
• Consistency across multiple sources

Retrieval Visualization

Visualize your retrieval pipeline to understand document ranking, relevance scores, and retrieval patterns. This helps identify:

• Retrieval bottlenecks
• Document relevance distribution
• Query-document matching patterns
• Performance optimization opportunities

Chunk Optimization

Get data-driven recommendations for optimal chunking strategies based on your content type and use case:

• Optimal chunk size recommendations
• Overlap configuration suggestions
• Splitting strategy analysis
• Performance impact predictions

Prompt Analysis

Analyze and optimize your prompts for better RAG performance:

• Prompt effectiveness scoring
• Context window optimization
• Instruction clarity analysis
• Best practice recommendations

Integration Guides

Framework-specific integration guides:

• LangChain: Native integration with LangChain chains and agents
• LlamaIndex: Plugin for LlamaIndex query engines
• Haystack: Custom component for Haystack pipelines
• Semantic Kernel: Connector for Microsoft Semantic Kernel

Support

Need help? We're here for you:

• Email: support@ragleaf.dev
• Discord: Join our community server
• GitHub: Report issues and contribute
• Contact: Contact form