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 long-term associative memory for LLMs by building a persistent concept graph with vector embeddings and semantic links.
Provides a self-hosted persistent memory layer for AI agents using semantic search and a local file vault.
$claude mcp add strata -- npx -y strata-mcp-server
Provides a local, chronological memory layer that allows AI clients to query and recall user-curated, dated memory banks via MCP.
$claude mcp add localkeep -- uvx localkeep
Provides self-hosted, cross-device long-term memory for AI agents using hybrid semantic and keyword search.
Lets AI assistants search a semantic knowledge graph, explore concept relationships, and check grounding confidence behind claims.
Provides distilled long-term memory for AI agents using meaning-based recall and evidence-gated writing to prevent memory poisoning.
Gives Claude Code persistent memory so sessions can save, search, and restore decisions, facts, and conversation chunks.
Provides a local-first, source-traceable memory layer for AI agents that remembers decisions and context across sessions without using an LLM at ingest.
$claude mcp add midas -- uvx midas-memory-mcp
Provides persistent, project-isolated memory for AI coding assistants using LLM-based fact extraction and semantic search.
Provides local, source-first memory for LLM agents using SQLite with hybrid FTS5 and vector search capabilities.
Provides local, versioned, and project-scoped memory for AI agents to maintain decision history and context across sessions.
Provides a local, verifiable memory layer for AI agents using SQLite to store, retrieve, and distill facts, patterns, and principles.
Provides AI coding agents with a shared project memory for decisions, code context, and persistent session continuity.
Provides a shared, versioned memory for AI tools to synchronize projects, personal context, and tasks across different MCP clients.
$claude mcp add --transport http innernet-mcp https://innernet.live/api/mcp
Provides AI agents with a local-first, durable, and inspectable memory layer using SQLite FTS5 for persistent context and evidence-aware verification.
$claude mcp add llm-kosh -- uvx llm-kosh
Provides an embedded fact store that tracks revisions to identify when remembered information has been superseded by newer facts.
$claude mcp add nanomem -- uvx nanomem
Provides a local-first memory server for AI assistants to store and retrieve structured research using BM25 search and portable files.
$claude mcp add dense-knowledge-mcp -- uvx dense-knowledge-mcp
Provides private on-device memory and retrieval for AI assistants using offline vector and lexical search.
$claude mcp add telys-sdk -- uvx telys
Provides a governed memory store for AI agents featuring signed provenance, policy-gated recall, and cryptographic erasure proofs.
$claude mcp add heartwood-memory -- uvx heartwood-memory
Provides AI agents with a local-first long-term memory system to store conversations, retrieve facts, and maintain continuity across sessions.
$claude mcp add hippocampus-mcp -- uvx hippocampus-mcp