MCP servers for memory & vectors
109 of the 8,106 servers we have read are placed here — embedding stores and agent memory. Placed by a stated rule over what each one declares it reaches and what it says it does, never by a classifier. Every row on this page, not a sample.
109 of the 8,106 MCP servers we have read are placed under memory & vectors. The most starred is iwe with 1,790 stars, and 3 are named as a vendor’s official server.
- iwe 1,790 stars
- mcp-server-qdrant 1,553 stars
- mcp-knowledge-graph 886 stars
- context-portal 767 stars
- tokensave 675 stars
Every server placed in memory & vectors
Markdown knowledge graph — LSP for your editor, CLI + MCP memory for your AI agents
No credential named
LISTED · ★ 1790An official Qdrant Model Context Protocol (MCP) server implementation
qdrantNeeds QDRANT_API_KEY
LISTED · ★ 1553MCP server enabling persistent memory for Claude through a local knowledge graph - fork focused on local development
No credential named
LISTED · ★ 886Context Portal (ConPort): A memory bank MCP server building a project-specific knowledge graph to supercharge AI assistants. Enables powerful Retrieval Augmented Generation (RAG) for context-aware development in your IDE.
No credential named
LISTED · ★ 767The most comprehensive code intelligence MCP server for AI coding agents. 40+ tools, 30+ languages, 9 agent integrations. Pre-indexed semantic knowledge graphs for instant code understanding — fewer tokens, fewer tool calls, 100% local.
No credential named
LISTED · ★ 675A Model Context Protocol (MCP) server implementation that provides database capabilities for Chroma
chromaNeeds CHROMA_API_KEY · CHROMA_CUSTOM_AUTH_CREDENTIALS
STALE · ★ 601Caura (formerly MemClaw) — governed shared memory for AI agent fleets. Multi-agent, multi-tenant, MCP-native. Trust tiers, keystone policies, audit trails, knowledge graph, self-improving retrieval. Apache 2.0.
No credential named
LISTED · ★ 504Your First LLM-Wiki Conversation Knowledge Base
beevercredentialgooglejina+1Needs BEEVER_MCP_KEY · CREDENTIAL_MASTER_KEY · GOOGLE_API_KEY · JINA_API_KEY · WEAVIATE_API_KEY
LISTED · ★ 451A knowledge graph server that uses the Model Context Protocol (MCP) to provide structured memory persistence for AI models.
No credential named
LISTED · ★ 354A graph DB-based MCP memory server for coding agents with intelligent relationship tracking
memorygraphNeeds MEMORYGRAPH_API_KEY
LISTED · ★ 255Model Context Protocol Servers for Milvus
No credential named
LISTED · ★ 245NeuralMemory stores experiences as interconnected neurons and recalls them through spreading activation, mimicking how the human brain works. Instead of searching a database, memories are retrieved through associative recall - activating related concepts until the relevant memory emerges.
No credential named
LISTED · ★ 242Persistent memory for AI coding agents
No credential named
LISTED · ★ 225next-generation AI memory infrastructure (powered by mem0 and graphiti)
No credential named
LISTED · ★ 170Open-source cognitive coprocessor with active memory for AI agents — persistent recall, semantic search, overnight dreaming, verified facts, encrypted USB sync. MCP server; works with Claude, ChatGPT, and any local LLM. Built by one maker and his agents.
No credential named
LISTED · ★ 158MCP server for persistent context management in AI coding assistants
Needs MCP_CHARS_PER_TOKEN
LISTED · ★ 136A vector embedding-based code semantic search tool with MCP server and multi-model integration. Can be used as a pure CLI tool. Supports Ollama for fully local embedding and reranking, enabling complete offline operation and privacy protection for your code repository
No credential named
LISTED · ★ 121Self-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.
googleqdrantNeeds GOOGLE_API_KEY
LISTED · ★ 111Needle MCP Server for easy RAG.Long-term memory for LLMs.
needleNeeds NEEDLE_API_KEY
STALE · ★ 104A memory system using mem0 for AI applications. Enables long-term memory for AI agents as a drop-in MCP server.
mem0openaisupabaseNeeds MEM0_API_KEY · OPENAI_API_KEY · SUPABASE_KEY
LISTED · ★ 96Context engine for large codebases, exposed through MCP. Gives AI coding agents precise repository context; benchmarked at frontier-agent quality with ~25x lower model cost and 45% fewer tokens with semantic search.
codealiveNeeds CODEALIVE_API_KEY
LISTED · ★ 90A MCP server that enables Claude to discover and call any API endpoint through semantic search. Intelligently chunks OpenAPI specifications to handle large API documentation, with built-in request execution capabilities. Perfect for integrating private APIs with Claude Desktop.
No credential named
STALE · ★ 82AI Memory
No credential named
LISTED · ★ 77Connect your Pinecone projects to Cursor, Claude, and other AI assistants
pineconeNeeds PINECONE_API_KEY
LISTED · ★ 72Semantic search for Hex documentation, right in your editor ✨
No credential named
STALE · ★ 70Graph-based long-term memory skill for AI (LLM) coding agents — faster context, fewer tokens, safer refactors
No credential named
LISTED · ★ 68Markdown vector search MCP server for Claude Code. Natural language search for markdown files using multilingual-e5-small embeddings.
No credential named
LISTED · ★ 64Outsource your understanding. Capture every human-AI decision as a step in a trajectory; grade trajectories retroactively when outcomes land (deal closed, sprint shipped). Hooks + MCP server for Claude Code, Qdrant + FastEmbed local-first.
qdrantNo credential named
LISTED · ★ 59Deep code indexing MCP server for AI agents. 25 tools: hybrid FTS5 + embedding search, call graphs, git blame/hotspots, build system analysis. Multi-repo workspaces, GPU-accelerated semantic search, 10 languages via tree-sitter. Fully local, zero cloud dependencies.
voyageNeeds VOYAGE_API_KEY
LISTED · ★ 58🧠 𝑴𝒆𝒎𝒐𝒓𝒚-𝑷𝒍𝒖𝒔 is a lightweight, local RAG memory store for MCP agents. Easily record, retrieve, update, delete, and visualize persistent "memories" across sessions—perfect for developers working with multiple AI coders (like Windsurf, Cursor, or Copilot) or anyone who wants their AI to actually remember them.
googleNeeds GOOGLE_API_KEY
STALE · ★ 56Persistent memory, semantic search, and shared project context for AI agents with traceable sources and scoped access. Carry decisions, lessons, and intent across sessions and tools.
contextstreamNeeds CONTEXTSTREAM_API_KEY
LISTED · ★ 44MCP server for Chroma
chromaNo credential named
STALE · ★ 41Persistent, branch-aware workflow state memory MCP server for AI coding assistants. Tracks tasks, accepted decisions, and active blockers to prevent session context bloat and speed up development.
No credential named
LISTED · ★ 40Persistent visual cache for LLM-driven software development. Caches screenshots using perceptual hashing, vector search, and AX trees to prevent token overhead and visual hallucination loops.
No credential named
LISTED · ★ 39Temporal memory system for AI assistants with human-like forgetting curves. All data stored locally in human-readable formats: JSONL for short-term memory, Markdown (Obsidian-compatible) for long-term. Memories naturally decay unless reinforced. Features knowledge graphs, smart prompting, and MCP server integration for Claude.
No credential named
LISTED · ★ 37Local FAISS vector store as an MCP server – Agent Memory, drop-in local semantic search for Claude / Copilot / Agents.
No credential named
LISTED · ★ 36Retrieval + inference offload for AI coding agents.
No credential named
LISTED · ★ 32Hosted persistent memory for AI agents over MCP. Tell it a recalled memory helped or misled, and it re-ranks what comes back next. Shared rooms for multi-agent work. One key or OAuth across Claude Code, Cursor, VS Code and ChatGPT. MIT, in the official MCP Registry.
mnemoverseNeeds MNEMOVERSE_API_KEY
LISTED · ★ 27The open retrieval layer for AI coding agents. Indexes code, docs, legal, research, data — 24 parsers (incl. EPUB, DOCX, ODT, MIDI, audio), FTS5 + semantic search, knowledge graph. Serves surgical context via MCP. Open source, local, free.
No credential named
LISTED · ★ 26MCP server for enabling persistent knowledge storage for Claude through a knowledge graph with multiple storage backends and fuzzy search
No credential named
LISTED · ★ 26Persistent cognitive memory for AI agents — drop-in MCP server for Claude Code, Cursor, and Windsurf with temporal decay, contradiction detection, a knowledge graph, and autonomous consolidation, backed by the YantrikDB Rust engine.
yantrikdbNeeds YANTRIKDB_API_KEY · YANTRIKDB_TOKEN
LISTED · ★ 25MCP Server for ChromaDB integration into Cursor with MCP compatible AI models
No credential named
STALE · ★ 25MCP server designed to help you search and analyze your photo library (iCloud - Apple photos)
No credential named
STALE · ★ 24🧠 High-performance persistent memory system for Model Context Protocol (MCP) powered by libSQL. Features vector search, semantic knowledge storage, and efficient relationship management - perfect for AI agents and knowledge graph applications.
libsqlNeeds LIBSQL_AUTH_TOKEN
STALE · ★ 24A simple MCP server that stores and retrieves memories from multiple LLMs.
No credential named
LISTED · ★ 22Stop AI Coding from forgetting. A knowledge graph–driven memory layer for LLMs (ChatGPT, Claude, Codex, DeepSeek, Gemini), enabling persistent long-term memory beyond context window limits. Build smarter AI agents with structured context, better consistency, and scalable multi-step reasoning across complex coding workflows.
No credential named
LISTED · ★ 20MCP server with semantic search + knowledge graph for Claude Code, Cursor, and Copilot
No credential named
LISTED · ★ 20Code Intelligence Engine — indexes your codebase and gives AI assistants deep understanding via MCP (semantic search, call graphs, 20+ tools)
No credential named
LISTED · ★ 19Self-hosted MCP server + dashboard giving Claude Code, Cursor & Hermes agents shared long-term memory over a git-backed markdown vault. Agent-native, Obsidian-compatible second brain — markdown RAG, no vector DB.
No credential named
LISTED · ★ 17Local-first agent memory MCP - cross-agent brain with token-saving file diff cache. By Synapse Arrows.
jwtNeeds JWT_SECRET
LISTED · ★ 15Never lose context again - persistent memory management system for AI-powered workflows across multiple tools
No credential named
STALE · ★ 14The Aura Emotion AI system has chroma with a local embedding model, memvid qr code mp4 infinite memory, brainwave and neurochemical simulations, sociobiological reasoning, autonomous subsystem processing with a Gemini flash model so the main model is less taxed, is a MCP client with adaptive tool learning and MCP server.
chromaNo credential named
LISTED · ★ 13Living memory for AI agents with an immune system. Episodes compress into identity. Zero dependencies, works with any client.
No credential named
LISTED · ★ 12XMemo: user-owned Memory OS for AI agents. Public client (@xmemo/client), MCP distribution, Skills and plugins.
xmemoNeeds XMEMO_KEY
LISTED · ★ 12MCP server giving AI coding agents offline, token-efficient persistent memory — recall decisions, patterns, and bugs across sessions with 68.5% fewer tokens than re-reading context. No LLM, no server; works with 15+ agents.
No credential named
LISTED · ★ 12Have LLMs save memories about you locally on your own hardware
No credential named
STALE · ★ 12A self-hostable, open-source, semantically-searchable Agent Skills registry delivered over MCP, with a three-tier progressive disclosure architecture.
qdrantworkersNeeds QDRANT_API_KEY · WORKERS_AI_API_TOKEN
LISTED · ★ 10High-performance MCP server for AI coding agents: persistent knowledge graph, blast radius analysis, semantic search. Built in Rust.
No credential named
LISTED · ★ 9knowledge graphs for AI
No credential named
LISTED · ★ 9MCP servers with local semantic search over React & React Native docs. Runs fully offline — no API key required.
No credential named
LISTED · ★ 8Shared Memory MCP server for agentic teams - solving coordination tax with 6x token efficiency
No credential named
LISTED · ★ 8Agentic OS that runs on any Linux — curated knowledge API for AI agents: MCP tools, skill packs, semantic search. Works with Claude Code, Cursor, Cline, Windsurf.
No credential named
LISTED · ★ 7Official Aiqbee AI developer tools and integrations. Connect AI assistants to your Aiqbee knowledge graph via MCP.
No credential named
LISTED · ★ 7MCP that brings persistent agent memory to Kiro, Claude Desktop, Cursor, and any MCP client — powered by agent-memory-daemon 🧠
No credential named
LISTED · ★ 6A vendor-agnostic cognitive persistence layer for AI agents. Eliminate the "repetition tax" by transporting your context, preferences, and history across sessions. Features an auto-adaptation engine that syncs global instructions to ensure operational cohesion and optimize token usage across any LLM or multi-agent workflow.
No credential named
LISTED · ★ 6Local MCP hub for personal data — private, cross-platform, agent-ready.
No credential named
LISTED · ★ 6Basic knowledge base mcp that can convert pdf files and make them searchable (using non-semantic search)
No credential named
LISTED · ★ 5MCP for managing tokens for team of agent or single one
No credential named
LISTED · ★ 5local-first temporal knowledge graph
alicebobNeeds ALICE_TOKEN · BOB_TOKEN
LISTED · ★ 5Distill AI coding sessions (Claude Code / Codex / Cursor) into weekly reports, ADRs, incident reviews and a knowledge graph — local-first, evidence-linked.
No credential named
LISTED · ★ 4Adaptive memory system for AI applications. Learns which retrieval strategies work for your data, scores results using cognitive science models, builds a knowledge graph automatically, and validates every parameter change against real query history before adopting it. Patent pending.
No credential named
LISTED · ★ 4simple memory mcp server with custom memory location
No credential named
STALE · ★ 4simple memory mcp server with custom memory location
No credential named
STALE · ★ 4MCP server for SkillsMP API - Search skills using keyword and AI semantic search
skillsmpNeeds SKILLSMP_API_KEY
LISTED · ★ 3Semantic search over 4.6 million text chunks from 20,000 classical philosophy and humanities works. Built for AI agents using FastMCP.
No credential named
LISTED · ★ 3A local-first code retrieval MCP server for AI coding agents. Hybrid keyword + vector search over Graphify-indexed codebases, with domain-aware ranking profiles.
No credential named
LISTED · ★ 3MCP server for Christian scholarship: BSB/WEB scripture, Greek/Hebrew word data, cross-references, patristic texts, and semantic search.
No credential named
LISTED · ★ 3Local-first second brain with semantic search. Stores notes as Markdown files and indexes them with a local ML model — no cloud, no API keys.
No credential named
LISTED · ★ 3Brain-inspired persistent memory for AI coding assistants. Hybrid BM25+vector search, ACT-R activation scoring, FadeMem decay, knowledge graphs, multi-agent coordination. Successor to claude-memory.
No credential named
LISTED · ★ 3跨工具共享记忆 + 主动建议的零依赖 MCP Server。Claude Code / Kimi Code / Cline / Cursor / Proma 共享同一份主动记忆,记住你教过的一切,并在合适时机主动开口提醒(含桌面通知 daemon)。npm @proactive-agent/mcp 0.9.2 · 本地优先 · 防投毒 · UMP 互操作
memoryNeeds MEMORY_LLM_API_KEY
LISTED · ★ 2A Model Context Protocol server that retrieves relevant code snippets or documents to help generating pymilvus code.
milvusopenaiNeeds MILVUS_TOKEN · OPENAI_API_KEY
STALE · ★ 2MCP server giving AI agents persistent memory: Markdown files as source of truth, hybrid BM25 + embedding search, typed graph relations. Self-hosted, single Docker image.
teamNeeds TEAM_API_KEY
LISTED · ★ 2Intelligent MCP (Model Context Protocol) router that selects tools via embeddings, semantic search, and RAG over a Qdrant vector database. Python · FastAPI · Qdrant · Docker.
qdrantNo credential named
LISTED · ★ 2Universal MCP server for document CRUD across formats (TXT, JSON, YAML, CSV, XML, DOCX, XLSX, PPTX, PDF) — with versioning, sandboxed multi-root workspace, full-text search, batch ops, and optional semantic search.
No credential named
LISTED · ★ 2Independent CLI benchmark harness for agent-memory backends (MemPalace, Mem0, Zep, OpenViking); publishes raw eval logs.
mem0memtrustopenvikingzepNeeds MEM0_API_KEY · MEMTRUST_JUDGE_API_KEY · OPENVIKING_API_KEY · ZEP_API_KEY
LISTED · ★ 2一个专为AI助手设计的记忆管理系统,实现了完整的 Model Context Protocol (MCP) 服务器,帮助AI助手存储、检索和管理用户的重要信息。
No credential named
STALE · ★ 1Local-first, reviewable memory layer for Codex with MCP, lifecycle hooks, and memory governance.
No credential named
LISTED · ★ 1Chartbrew + AI agents. MCP server exposing Chartbrew's documented API: teams, connections, datasets, dashboards, charts, live queries, and secure embedding. TypeScript · stdio · restricted/unrestricted tool modes.
chartbrewNeeds CHARTBREW_API_KEY
LISTED · ★ 1MCP server for Zep Ai
zepNeeds ZEP_API_KEY
STALE · ★ 1Production-hardened MCP memory server for AI agents: hybrid BM25+FAISS search, circuit breaker, SLO tracking, Prometheus metrics, BND boundary evaluation. Gives AI persistent long-term memory that survives across sessions.
No credential named
LISTED · ★ 1A production-grade semantic search server for food recipes — built for AI agents using the Model Context Protocol (MCP). Search across 50,000+ recipes with hybrid dense + sparse retrieval and cross-encoder reranking.
No credential named
LISTED · ★ 1Weaviate MCP Client for FlowMasters AI Agency
weaviateNeeds WEAVIATE_API_KEY
STALE · ★ 1A local-first, generic persistent memory layer for MCP agents, with semantic search, revisions, provenance, temporal recall, and relationships.
No credential named
LISTED · ★ 1Python Chroma MCP
chromaNo credential named
STALE · ★ 1Repository knowledge graph for Cursor and Claude Code. Trace callers, callees, symbol definitions, and dependencies to understand unfamiliar code and make better-informed changes.
No credential named
LISTED · ★ 1Living memory system for AI assistants — SQLite + MCP with decay, learning, and knowledge graph
memorysqliteNeeds MEMORY_API_TOKEN
LISTED · ★ 1MCP server for a compliance knowledge graph: 686 frameworks, 21,696+ controls, 310K+ cross-framework mappings. Node and Python implementations over stdio, plus a hosted endpoint.
No credential named
LISTED · ★ 0Local semantic search over React Native docs for AI agents — no API key
No credential named
LISTED · ★ 0MCP server over a cited knowledge graph on the Laws of the Game and world football competitions. Remote streamable-HTTP, 8 tools, no auth.
No credential named
LISTED · ★ 0Self-installing MCP memory server using Qdrant Embedded DB. Zero-config setup for Claude Code, Codex, Cursor and more. Compatible with any MCP-capable client.
qdrantNo credential named
LISTED · ★ 0MCP server over a cited knowledge graph on testing, benchmarking and auditing AI agents. Remote streamable-HTTP, 8 tools, no auth.
No credential named
LISTED · ★ 0MCP server over a cited knowledge graph on Spain's contributions to science, exploration and culture. Remote streamable-HTTP, 8 tools, no auth.
No credential named
LISTED · ★ 0Lightweight local-first persistent memory for coding agents and MCP clients.
No credential named
LISTED · ★ 0MCP server for semantic search with Qdrant vector database
qdrantNeeds QDRANT_API_KEY
STALE · ★ 0LanceDB MCP Server nodejs
No credential named
STALE · ★ 0ByteRover Team Memory public documentation modeled after the MiroFish docs pattern.
No credential named
LISTED · ★ 0Hjarni MCP Server — Give your AI long-term memory. Note-taking app with built-in MCP server for Claude and ChatGPT.
No credential named
LISTED · ★ 0enhanced memory mcp from official code (fuzzy search, etc)
No credential named
STALE · ★ 0Art provenance MCP server — 282K-node knowledge graph, cited answers, named custody gaps. Free open beta at provenio.art
No credential named
LISTED · ★ 0
Common questions about MCP servers for memory & vectors
- What are the most popular MCP servers for memory & vectors?
- By GitHub stars: iwe (1,790 stars), mcp-server-qdrant (1,553 stars), mcp-knowledge-graph (886 stars), context-portal (767 stars), tokensave (675 stars). Stars measure popularity, not quality, so read what each one reaches and how it is maintained.
- What do MCP servers for memory & vectors connect to?
- Among those placed here, the most common things they say they reach are qdrant (7 servers), chroma (4 servers), google (3 servers), mem0 (2 servers), memory (2 servers), openai (2 servers), weaviate (2 servers), zep (2 servers).
- Which MCP servers for memory & vectors need a credential?
- 30 of 109 name a credential in their source; the most common are GOOGLE_API_KEY (3), QDRANT_API_KEY (3), MEM0_API_KEY (2), OPENAI_API_KEY (2), WEAVIATE_API_KEY (2). A server with no credential listed has not been shown to need none; check its README.
- Are there official MCP servers for memory & vectors?
- Yes: mcp-server-qdrant, chroma-mcp and caura are named as a vendor’s own server.
- How are servers placed under memory & vectors?
- By a stated rule over what each one declares it reaches and what it says it does, never by a classifier. A server can sit in no category when no rule fired, which does not mean it belongs nowhere.