MCP servers for long-term memory and vector search: store notes, embeddings and facts so an AI assistant remembers across sessions. 799 servers in this category.
Provides local-first personal memory for AI agents using a SQLite store where agents propose information for human confirmation.
$claude mcp add memory-hive-mcp -- uvx memory-hive-mcpProvides local-first personal memory for AI agents, allowing them to propose and search information stored in a local SQLite database.
$claude mcp add apiary-memory -- uvx apiary-memory
Provides a local-first SQLite episodic memory engine that allows AI coding agents to maintain project state and architectural decisions across different models.
Screens an AI agent's memory writes and reads through detectors and a declarative policy to block poisoning attacks.
Lets an AI client connect to and explore relational databases, Redis, MongoDB, Milvus, Elasticsearch, and Kafka through one MCP server.
Gives AI coding agents persistent local memory of decisions, solutions, facts and session summaries through MCP tools.
$claude mcp add total-agent-memory -- uvx total-agent-memory
Lets an AI client create, populate, list and delete vector collections and run semantic, lexical, and fused searches across them.
Provides local Markdown-based memory for LLM agents using BM25 retrieval and optional semantic reranking.
Provides a local memory store for coding agents that ensures every recalled fact includes a source and anchor to prevent hallucinations.
$https://github.com/menot-you/n-memory/releases/download/v0.1.2/nmemory-linux-x86_64.mcpb
Provides local-first episodic memory for AI agents to store and retrieve decisions, notes, and context across different sessions and models.
$claude mcp add cairn -- uvx cairn-remembers
Provides semantic memory for AI agents by indexing local documents and chat histories into a queryable FAISS vector store.
$claude mcp add memory-os-ai -- uvx memory-os-ai
Provides self-hosted persistent memory for AI agents using a knowledge graph and hybrid semantic search to maintain project context across sessions.
$claude mcp add maxgraph -- docker run -i --rm ghcr.io/rumemory/maxgraph:1.9.5
Provides durable AI memory that persists distilled facts across different sessions and subagents.
$claude mcp add khwan-mcp -- uvx khwan-mcp
Gives AI coding agents persistent, locally stored memory so they recall past decisions, preferences, and lessons across sessions.
$claude mcp add omega-memory -- uvx omega-memory
Lets an AI client search, organize, and maintain a Calibre e-book library through semantic search, RAG, and library management tools.
Provides a governed memory layer for multi-agent systems featuring role-based access, provenance, and temporal replay.
$claude mcp add agent-memory -- uvx memwright
Manage memory infrastructure for AI agents by creating embedders, rerankers, and LLMs, and ingesting memories for RAG and deep research.
Provides local long-term memory for Claude Code and other agents by storing and retrieving exact conversation history and command outputs.
$claude mcp add cogito-mem -- uvx cogito-mem
Provides AI agents with durable, searchable memory using hybrid semantic and keyword retrieval with autonomous consolidation.
$claude mcp add kirok-mcp -- uvx kirok-mcp
Provides shared long-term memory for AI agent fleets, allowing agents to record and retrieve insights across an organization.
$claude mcp add hivemind -- docker run -i --rm ghcr.io/monochrome-muncher/hivemind:v2.26.0