"Servers Offering MCP Prompts as Prompt Primitives" matching MCP servers:

Provides stateful prompt optimization using research-backed techniques like APE and OPRO, learning from historical performance data via a vector database. It enables users to automatically refine prompts, retrieve high-performing examples, and track performance analytics through iterative feedback.

将 pgvector 知识库检索能力封装为标准 MCP 服务器，支持 Tools、Resources 和 Prompts，使 Claude Desktop 等客户端可通过自然语言查询知识库。

Enables agents to run hybrid dense and BM25 search over a local folder of Markdown files, read and write notes, and trigger reindexing as the folder changes. It also injects the most relevant sections into each prompt automatically and runs entirely locally with a bundled embedding model.

Enables AI agents to parse PDFs and search local document context through multimodal extraction and vector-based retrieval, all exposed as MCP tools.

A local-first knowledge base for LLM coding agents that indexes repository documentation, concept ontology, and build targets into Qdrant and exposes retrieval as MCP tools (search, get, list sources, reindex).

A semantic search system for Claude Code transcript history, exposed as an MCP server so Claude Code can query its own past sessions.

Enhanced MCP knowledge graph memory server with cloud persistence and semantic search, acting as a drop-in replacement for the standard memory server.

Enables AI assistants to interact with a Qdrant vector database by exposing collection, point, vector, payload, snapshot, search, recommendation, discovery, and observability operations as MCP tools.

A local MCP memory server for Kiro CLI that provides persistent semantic memory across sessions using SQLite for storage, Ollama for local vector embeddings, and Obsidian as a human-readable sync target.

Keyless local RAG over your own documents (ChromaDB) with traceable citations, exposed as MCP tools and a FastAPI endpoint.

Self-hosted mem0 MCP server for Claude Code. Run a complete memory server against self-hosted Qdrant + Neo4j + Ollama while using Claude as the main LLM.

Semantic vector DB as an MCP server for Claude Code — SQLite + FTS5 + local ONNX embeddings. Gives Claude persistent, searchable project knowledge across conversations.

A fully-local, three-layer memory plugin for Claude Code with Chinese support, offering hybrid FTS5/trigram and HNSW vector search, self-iterating memory management, consolidation, and a web console.

Exposes semantic document retrieval from a ChromaDB collection of arXiv papers as a callable tool for any MCP-compatible client.

A high-performance, persistent memory system for the Model Context Protocol (MCP) providing vector search capabilities and efficient knowledge storage using libSQL as the backing store.

A flexible memory system for AI applications that supports multiple LLM providers and can be used either as an MCP server or as a direct library integration, enabling autonomous memory management without explicit commands.

An MCP server for storing and retrieving memories using Qdrant vector search, acting as a semantic memory layer.

Exposes infocepo.com infrastructure (LLM, ChromaDB, S3, Docker registry, service discovery) as MCP tools for AI agents.

An offline-first, governed memory and knowledge server for AI agents that provides Remember, Search, Update, and Forget operations with hybrid retrieval, semantic embeddings, and NID-based authentication. It can be used as an MCP server via stdio or Streamable HTTP, enabling agents to persist and query memories and wiki knowledge.

Enables users to index and retrieve local documents through semantic, BM25, or hybrid search, with tools for chunk reads and page images accessible from MCP clients such as Codex and ChatGPT web.

Semantic memory server for AI agent teams. Stores, searches, and retrieves knowledge across sessions using pluggable vector backends with local ONNX embeddings, exposed as an MCP server.

Enables semantic and hybrid search over documents and codebases backed by a Qdrant vector database, with collection management, document CRUD, metadata filtering, and AST-aware code indexing using local or cloud embedding providers. Supports natural language queries, incremental re-indexing of changed files, and custom configurable prompts for guided workflows.

Exposes a shared graph-based context engine as MCP tools for Claude, Copilot, and other AI agents, enabling knowledge ingestion, recall, search, and LLM-ready context assembly across sessions.

Turns any folder of PDFs, markdown, and text files into a local, queryable knowledge base exposed as an MCP server. Enables MCP-compatible agents to semantically search indexed documents, retrieve relevant passages with source and relevance scores, list or reindex documents, and inspect cache and token usage — instead of reading whole files into context.

Exposes LangChain and Anthropic Claude capabilities as tools for generating production-ready RAG systems, Supabase vector stores, and document ingestion pipelines. It enables users to instantly scaffold AI infrastructure and document processing code through natural language prompts in MCP-compatible clients.

A minimal RAG service that exposes a vector index for document retrieval via REST and MCP, allowing querying for relevant document chunks and returning a suggested LLM prompt.

Enables MCP clients such as Claude Code to give LLM agents long-term memory by exposing remember, recall and forget tools over stdio, backed by a hybrid BM25 and vector store that reinforces frequently used memories.

Enables storing and retrieving information using semantic search with Qdrant vector database. Acts as a memory layer for LLMs to persistently store and semantically search through information and metadata.

Provides long-term memory for AI agents through time-decay validity scoring, so stored facts, notes, and entities are automatically forgotten or flagged as stale instead of being recalled forever. Enables logging high-signal memories, searching them with freshness guidance, extracting daily timelines, and maintaining long-lived entity references across sessions.

Enables vector search and document retrieval, extraction and chunking of any file into Markdown, and private deep research generation through Vectorize pipelines. It connects MCP clients such as Claude, Cursor, and VS Code to these Vectorize APIs.