Intelligent Retrieval-Augmented Generation (RAG) system with domain-specific routing and semantic search across 1.5M+ chunked documents from 208+ sources across 8 knowledge domains.

• Domain-Specific Routing: QueryRecognizer auto-classifies queries across 8 knowledge domains
• 50-70% Search Optimization: Domain-focused retrieval reduces search scope
• <3ms Query Recognition: O(1) keyword-based domain classification
• 1.5M+ Indexed Chunks: From 208 sources across 8 domains
• Claude API Integration: Enhanced reasoning with domain-specific context
• Production Ready: Fully tested and deployed with EnhancedRAGAgent

1. Design & Graphics (13 sources) - Figma, GIMP, UI/UX design
2. DTF Printing & Business Automation (23 sources) - Direct-to-film, n8n automation
3. Legal & Compliance (16 sources) - Business formation, contracts, regulations
4. SaaS & Software Law - Specialized legal issues for software companies
5. Intellectual Property - Patents, trademarks, licensing
6. Fundraising & Venture Capital (16 sources) - Cap tables, SAFE agreements, pitch decks
7. E-Commerce (11 sources) - Shopify, payment processing, fulfillment
8. General (cross-domain fallback)

Fully Cloud-Compatible - This is an MCP server system for integration with Claude Code CLI and compatible AI agents. Not designed for local installation.

AI Agent (Claude, GPT, etc.) → EnhancedRAGAgent
                                    ↓
                            QueryRecognizer (domain classification)
                                    ↓
                            ChromaDB HTTP Client (localhost:8001)
                                    ↓
                            Vector Database (1.5M chunks, 8 domains)

• Intelligent Query Routing - Automatically classifies queries across 8 domains
• Domain-Specific Search - 50-70% search scope reduction via focused filters
• Fast Recognition - <3ms domain classification with O(1) inverted index
• Semantic Search - ChromaDB vector database with 384-dim sentence embeddings
• Claude API Ready - Integrates with Claude for sophisticated AI reasoning
• Production Tested - Fully deployed, tested, and verified system

• EnhancedRAGAgent (rag_agent_enhanced.py) - Query router with ChromaDB integration
• QueryRecognizer (rag_agent_optimizer.py) - Domain classification engine
• System Prompts - 2 comprehensive prompts for LLM agents
• Domain Registry - Metadata for all 8 knowledge domains
• Diagnostic Tools - ChromaDB metadata inspection and testing

• DEPLOYMENT_STATUS.md - Current system status and test results
• DOMAIN_MONETIZATION_ANALYSIS.md - Market analysis for domain-specific SaaS
• RAG_AGENT_INTEGRATION_GUIDE.md - Developer integration guide
• RAG_OPTIMIZATION_SUMMARY.md - Technical optimization details
• test_rag_deployment.py - Comprehensive test suite

For using with Claude Code CLI or similar AI agents:

• ChromaDB HTTP server running at localhost:8001
• Python 3.12+
• chromadb, sentence-transformers, langchain libraries

Note: This is a production RAG system. Local installation/setup is not supported for public use. The system is designed for integration with cloud-based AI agents.

MIT License - see LICENSE

✅ Production Ready

• Query Recognition: All 10 test queries passed ✓
• Domain-Specific Retrieval: 3/3 tests passed ✓
• Error Handling: Comprehensive error recovery ✓
• Integration: Ready for Claude API integration ✓

Built with Claude Code, ChromaDB, sentence-transformers, LangChain, and Anthropic Claude

⭐ Status: ✅ FULLY DEPLOYED & TESTED | Knowledge Base: 208 sources, 1.5M+ chunks | Last Updated: November 17, 2025 | Generated with: Claude Code