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 a vendor-neutral memory layer for AI agents using three-tier short, mid, and long-term memory horizons.
Saves git-based session handoffs so the next coding agent, in any MCP tool or machine, starts with what was done.
$claude mcp add remembra -- uvx remembra-mcp
Provides persistent, shared organizational memory for AI agents using a local SQLite knowledge graph and hybrid semantic search.
Turns markdown notes into a queryable knowledge graph to provide persistent, semantic memory for AI assistants.
$claude mcp add memograph -- uvx memograph
Provides long-term memory for AI agents by storing facts, user profiles, and task experiences across sessions.
Provides AI agents with persistent long-term memory, a temporal knowledge graph, and outcome-based learning to prevent session data loss.
Provides local, private, long-term memory for AI clients by storing and searching verbatim notes, files, and conversations.
$claude mcp add engram -- uvx engram-chateau
Provides a multi-layer local memory system for AI agents using SQLite to store facts, episodic memories, knowledge graphs, and project context.
$claude mcp add mcp-ariel-memory -- npx -y mcp-ariel-memory
Provides MCP hosts with long-term agent memory, cross-session project recall, and durable conversation recording with speaker attribution.
$https://github.com/reshuibuduo/tmcra-mcp-server/releases/download/v1.0.0-rc.1/tmcra-mcp-server-1.0.0-rc.1.mcpb
Provides local-first agent memory that allows AI coding agents to source, date, and bind information to prevent hallucinations.
$claude mcp add lbrain-coding-agents -- uvx lbrain-coding-agents
Gives AI coding agents a shared, self-hosted knowledge graph for storing and retrieving context across sessions and tools.
Lets an AI client run semantic and hierarchy-aware searches over a Qdrant vector database built from Git, Confluence, JIRA, and local files.
Persistent memory server for Claude Code and OpenCode that stores advice and promotes or demotes it based on scored outcomes.
Gives Google ADK agents durable Redis sessions, semantic long-term memory, vector search tools, and semantic caching.
Provides a persistent, versioned file-system workspace for AI agents to store, search, and recall knowledge across sessions.
$claude mcp add sayou -- uvx sayou
Provides long-term memory for AI agents using knowledge graphs, provenance tracking, and cascading stale-fact detection.
Provides durable hybrid memory for AI agents using vector, lexical, temporal, and graph retrieval to maintain context across sessions.
Provides a deterministic, self-hosted long-term memory layer for AI agents to store and recall facts without using an LLM on the read or write path.
Provides a portable, local-first memory system that stores user preferences and decisions for MCP-compatible AI clients.
Provides local semantic search and agent memory by indexing over 20 document formats using a local ONNX model and LanceDB.