{"id":"w3-mcp-server-qdrant","name":"w3-mcp-server-qdrant","af_score":55.8,"security_score":42.2,"reliability_score":21.2,"what_it_does":"Provides a Python MCP server exposing tools to query a Qdrant vector database using text queries. It can auto-generate embeddings via an external Ollama server and supports advanced retrieval techniques such as query expansion (with RRF merging), HyDE, and LLM-based reranking.","best_when":"You want an agent-callable vector search tool that enriches queries via Ollama-generated embeddings and optionally uses additional LLM steps (expansion/rerank) to improve retrieval quality.","avoid_when":"You cannot run/allow the supporting services (Qdrant and Ollama) or you need robust built-in auth/rate limiting controls at the MCP server boundary.","last_evaluated":"2026-04-04T21:39:53.021841+00:00","has_mcp":true,"has_api":false,"auth_methods":["MCP stdio (stdio transport implied by usage examples)","Qdrant API key via QDRANT_API_KEY (optional per README)"],"has_free_tier":false,"known_gotchas":["Advanced options (expand_query/use_hyde/rerank) may significantly increase latency since they require additional LLM calls and parallel searches.","If Ollama or Qdrant is unavailable, the server may hang or fail; README suggests manual health checks but does not describe MCP-level retry semantics.","No pagination mechanism is described; limit is a single parameter (no continuation token).","Reranking/query-expansion depend on OLLAMA_RERANK_MODEL being available; misconfiguration will break those paths."],"error_quality":0.0}