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 persistent long-term memory for coding agents using associative stores, canonical facts, and knowledge graphs.
$claude mcp add pseudolife-mcp -- uvx pseudolife-mcp
Provides agent memory using SQLite or PostgreSQL that validates facts and refuses to answer unsupported queries instead of guessing.
$claude mcp add re-call -- uvx recall-rag
Provides a local-first memory engine for AI applications to manage evidence, world objects, and provenance through controlled reads and writes.
$claude mcp add memoweft -- npx -y @memoweft/mcp-server
Provides a local-first temporal knowledge graph that allows AI agents to retrieve project decisions through read-only tools and human-reviewed writes.
$claude mcp add graphiti-local -- uvx graphiti-local
Provides a neuro-inspired long-term memory layer for AI agents using semantic graphs and asynchronous background consolidation.
$claude mcp add smriti-memory -- uvx smriti-memcoreProvides a neuro-inspired long-term memory engine for AI agents using semantic graphs and asynchronous consolidation.
$claude mcp add smriti-memcore -- uvx smriti-memcore
Provides self-hosted shared memory for AI agent fleets using hierarchical rooms, halls, and layers for scoped semantic retrieval.
$claude mcp add fleet-memory -- npx -y fleet-memory-mcp
Provides persistent, searchable shared memory for AI coding agents across different clients and sessions using hybrid search and a typed-edge graph.
Provides a persistent, Mem0-compatible memory layer for AI agents using hybrid retrieval and LLM fact extraction.
$claude mcp add --transport http deepmem https://deepmem.dev/mcp
Gives AI agents a durable memory layer to capture episodes, curate concepts, and recall context with temporal facts and semantic search.
Gives an AI agent persistent, governed memory with hybrid BM25F and vector recall, audited writes, and drift detection.
Gives AI clients a local SQLite knowledge graph with FTS5 hybrid search, session and task tracking, and cross-machine bridge sync.
Provides a behavioral memory layer that learns user preferences and project context across different AI clients.
$claude mcp add magnet-gateway -- uvx agent-magnet
Provides a local, long-term memory for AI agents that stores conversations word for word on your own PC without using LLM tokens for saving.
$claude mcp add turnbook -- uvx turnbookProvides local long-term memory for AI agents by storing conversation transcripts word for word in a local SQLite database.
$claude mcp add graph-mind -- uvx graph-mind-memory
Provides an append-only ledger of sourced, dated, and confidence-scored claims for AI agents to use as shared memory.
$claude mcp add particles-engine-py -- uvx linkedparticles
Provides a reconciling memory layer for AI agents that manages fact updates, maintains history, and supports human approval for high-stakes changes.
$claude mcp add vayl -- uvx vayl-mcp
Provides an episode-first, zero-latency long-term memory layer for local LLMs and AI coding assistants using hybrid SQLite search.
$claude mcp add orma -- uvx orma-ai
Provides AI agents with local-first, tamper-evident memory and governance using a hash-chained SQLite ledger.
$claude mcp add arkhive -- uvx arkhive-mcp
Provides AI clients with a persistent, shared memory engine using a knowledge graph to recall facts and preferences across different sessions.
$claude mcp add spomory -- uvx spomory