Qdrant MCP Server

# Examples Practical examples demonstrating Qdrant MCP Server usage. ## Prerequisites Before running examples: ```bash # Start services docker compose up -d # Pull embedding model docker exec ollama ollama pull nomic-embed-text ``` Configure MCP server as described in [main README](../README.md). ## Available Examples ### 🎯 [Basic Usage](./basic/) **Start here** - fundamental operations - Creating collections - Adding documents - Semantic search - Resource management **Time:** 5-10 minutes | **Difficulty:** Beginner --- ### 🔀 [Hybrid Search](./hybrid-search/) Combine semantic and keyword search for better results - Understanding hybrid search benefits - Creating hybrid-enabled collections - Comparing semantic vs hybrid search - Best practices for technical content **Use cases:** Technical docs, product search, legal documents, code search **Time:** 15-20 minutes | **Difficulty:** Intermediate --- ### ⚡ [Rate Limiting](./rate-limiting/) Automatic rate limit handling for batch operations - Configuring provider rate limits - Batch document processing - Exponential backoff retry - Monitoring and troubleshooting **Use cases:** High-volume ingestion, free tier optimization, production reliability **Time:** 10-15 minutes | **Difficulty:** Beginner to Intermediate --- ### 📚 [Knowledge Base](./knowledge-base/) Searchable documentation system with metadata - Structuring documents with rich metadata - Organizing by team, topic, difficulty - Filtering searches by categories - Scaling and maintenance **Use cases:** Company docs, help centers, internal wikis, education **Time:** 15-20 minutes | **Difficulty:** Intermediate --- ### 🔍 [Advanced Filtering](./filters/) Complex search filters with boolean logic - Multiple filter conditions (AND, OR, NOT) - Filtering by categories, ratings, availability - Range filters for numeric values - E-commerce search patterns **Use cases:** Product catalogs, inventory, content filtering, access control **Time:** 20-30 minutes | **Difficulty:** Intermediate to Advanced ## Learning Path ``` Basic → Hybrid Search → Rate Limiting → Knowledge Base → Advanced Filtering ``` ## Common Patterns | Pattern | Metadata Structure | Use Case | | -------------------- | -------------------------------------------------- | --------------------- | | Content Organization | `category`, `topic`, `author`, `date` | Blogs, docs, articles | | E-commerce | `category`, `price`, `rating`, `in_stock`, `brand` | Product catalogs | | Access Control | `visibility`, `department`, `sensitivity` | Enterprise knowledge | | Temporal Data | `created_at`, `updated_at`, `status`, `version` | Versioned content | ## Tips 1. **Start Small** - Test with 5-10 documents before scaling 2. **Design Metadata First** - Plan fields, types, and filters 3. **Use Consistent IDs** - Choose a scheme (sequential, prefixed, semantic, UUIDs) 4. **Test Searches** - Validate semantic matching and filters 5. **Clean Up** - Delete test collections when done ## Troubleshooting | Issue | Solution | | ---------------------- | --------------------------------------------------- | | Collection exists | `Delete collection "name"` then recreate | | No search results | Check collection has documents, try without filters | | Unexpected results | Validate metadata and filter syntax | | "Collection not found" | Create collection first | | "Bad Request" | Check filter JSON syntax | | API errors | Verify provider API key and credits | ## Next Steps 1. Review [main README](../README.md) for full tool documentation 2. Apply patterns to your own use cases 3. Explore advanced features and configurations 4. Share your examples via [CONTRIBUTING.md](../CONTRIBUTING.md) ## Additional Resources - [Qdrant Documentation](https://qdrant.tech/documentation/) - [OpenAI Embeddings](https://platform.openai.com/docs/guides/embeddings) - [Cohere Embeddings](https://docs.cohere.com/docs/embeddings) - [Voyage AI](https://docs.voyageai.com/) - [Ollama](https://ollama.ai/docs) - [Model Context Protocol](https://modelcontextprotocol.io/)