Filter documents by metadata before ranking by vector similarity to enable production RAG and semantic search pipelines.
MIT
205,579 tools. Last updated 2026-06-17 07:46
"A local vector-based search engine for personal documents" matching MCP tools:
- Find exact matches for error codes, IDs, and names using BM25 keyword search over indexed documents. Use when precise term recall is required.MIT
- Search for similar documents by vector similarity, with optional scalar filters like age > 25.Apache 2.0
- Find relevant content across namespaces using natural language queries or vector similarity. Filter results by metadata or keywords for precise discovery.Apache 2.0
- Combine BM25 keyword search with vector ANN search in a single pass. Use for RAG when either semantic or keyword search alone is insufficient.MIT
- Search for documents by content and metadata criteria. Use content-based retrieval and property filters to narrow results.Apache 2.0
Matching MCP Servers
- Flicense-qualityBmaintenanceA CLI-first semantic code search tool with MCP integration for AI assistants, enabling semantic search, AST-aware parsing, and code analysis across 13 languages.Last updated47
- Flicense-qualityDmaintenanceEnables semantic search over local notes and documents using natural language queries. Supports multiple file types (Markdown, Python, HTML, JSON, CSV, text) with fast local embeddings and persistent ChromaDB vector storage.Last updated1
Matching MCP Connectors
Find local businesses on Google: name, address, phone, hours, ratings, and photos.
Cloudflare Workers MCP server: agent-workflow-engine
- Converts plain text documents into dense vector embeddings and upserts them into a Zvec collection for storage.Apache 2.0
- Set up a Redis 8 vector similarity index using HNSW to perform approximate nearest neighbor search on hash-stored documents with float32 vectors.MIT
- Start a browser-based authorization, poll for user approval, and store the returned MCP personal token.MIT
- Search markdown corpus with keyword or vector search on IP documents. Apply filters for energy, depth, project, or TODOs. Set result limit and minimum depth to retrieve targeted documents.MIT
- Retrieve relevant information from your local documents using combined keyword and vector search. Results include a relevance score.MIT
- Index workspace documents into vector storage for search and memory. Use during initial setup or to rebuild the entire index from scratch.AGPL 3.0
- Search the web using Brave Search for global content or Naver Search for Korean-specific results. Automatically selects the appropriate engine based on configured API keys.MIT
- Delete all indexed documents and vector data to perform a full factory reset. Requires confirmation to proceed.Elastic 2.0
- Retrieve relevant documents from your local knowledge base by combining exact keyword matching with semantic search. Results include relevance scores for ranking.MIT
- Retrieve privacy-filtered personal context from your local memory by searching for a specific topic. Use this to answer questions about your history, preferences, or stored information.MIT
- Search documents in Elasticsearch indexes with advanced filtering, pagination, and time-based sorting to find relevant information.MIT
- Ingest documents (PDF, DOCX, TXT, MD) into a vector database for semantic search. Supports updating existing documents and visual captioning for PDF figures.MIT
- Ingest documents (PDF, DOCX, TXT, MD) into a local vector database for semantic search. Supports re-ingestion to update existing content.MIT
- Search a document store's vector store to retrieve relevant documents based on a query.MIT