Skip to content

Best-of MCP topic

Best MCP Servers for AI Memory & Context

Top MCP servers for persistent memory, sequential reasoning, and context management. Make AI agents smarter.

20 servers · Updated 2026 · From 2805 reviewed

Top Picks

#1 Shodh Memory Scan found nothing
161 stars

Cognitive memory for AI agents with Hebbian learning, 3-tier architecture, and knowledge graphs. Single ~15MB binary, runs offline on edge devices.

#2 Memdocs Scan found nothing
1 stars

Git-native project memory for AI assistants with enterprise-grade audit compliance

Compatibility

Claude CodeFull support
Tested
CursorFull support
Tested
VS CodeFull support via Copilot
Tested
WindsurfFull support
Tested
Claude DesktopFull support
Tested

Install

Claude Code

claude mcp add memdocs -- uvx memdocs
#3 ApeRAG Scan found nothing
1,068 stars

Production-ready RAG platform combining Graph RAG, vector search, and full-text search. Best choice for building your own Knowledge Graph and for Context Engineering

All AI Memory & Context Servers

Codebase Memory

Scan found nothing
by DeusData

Persistent codebase knowledge graph. Survives session restarts and context compaction.

ai memory C stdio
687
View

Rust Docs MCP Server

Scan found nothing
by Govcraft

Provides up-to-date documentation context for a specific Rust crate to LLMs via an MCP tool, using semantic search (embeddings) and LLM summarization.

documentation Rust stdio
262
View

MCP Ragdocs

Scan found nothing
by hannesrudolph

An MCP server implementation that provides tools for retrieving and processing documentation through vector search, enabling AI assistants to augment their responses with relevant documentation context

ai memory TypeScript stdio
256
View

Mengram

Scan found nothing
by alibaizhanov

Human-like memory layer for AI agents with semantic, episodic, and procedural memory. Claude Code hooks (auto-save, auto-recall, cognitive profile). 29 MCP tools, knowledge graph, smart triggers, multi-user isolation. Python & JS SDKs.

ai memory Python stdio
108
View

Ragieai MCP Server

Scan found nothing
by ragieai

Retrieve context from your Ragie (RAG) knowledge base connected to integrations like Google Drive, Notion, JIRA and more.

ai memory JavaScript stdio
87
View

Ldclabs MCP Server

Scan found nothing
by ldclabs

Knowledge Graphs to enable memory persistence, knowledge evolution, explainable interaction.

ai memory Python stdio
claude mcp add kip -- uvx kip
56
View

Omega Memory

Scan found nothing
by omega-memory

Persistent memory for AI coding agents with semantic search, auto-capture, cross-session learning, and intelligent forgetting. 12 MCP tools, local-first.

ai memory Python stdio
claude mcp add omega-memory -- uvx omega-memory
41
View

Core

Scan found nothing
by omega-memory

Persistent memory for AI coding agents. #1 on LongMemEval. Local-first.

ai memory Python stdio
claude mcp add core -- uvx core
41
View

Mnemex

Scan found nothing
by prefrontal-systems

Temporal memory for AI with decay and reinforcement. Two-layer storage (JSONL + Markdown).

ai memory HTML stdio
29
View

Engram

Scan found nothing
by tstockham96

Intelligent agent memory with semantic recall, automatic consolidation, contradiction detection, and bi-temporal knowledge graph. 80% on LOCOMO benchmark using 96% fewer tokens than full-context approaches.

ai memory TypeScript stdio
claude mcp add tstockham96-engram -- npx -y engram
27
View

Documentation Extractor

Scan found nothing
by nirholas

Extract llms.txt from any docs site - Mintlify Docusaurus GitBook parser

documentation TypeScript stdio
claude mcp add extract-llms-docs -- npx -y extract-llms-docs
15
View

Neverinfamous Memory Journal MCP

Scan found nothing
by neverinfamous

A MCP server built for developers enabling Git based project management with project and personal…

ai memory TypeScript stdio
claude mcp add memory-journal-mcp -- npx -y memory-journal-mcp
11
View

Remembra

Scan found nothing
by remembra-ai

Persistent memory layer for AI agents with entity resolution, PII detection, AES-256-GCM encryption at rest, and hybrid search. 100% on LoCoMo benchmark. Self-hosted.

ai memory HTML stdio
6
View

Connapse

Scan found nothing
by Destrayon

Self-hosted knowledge backend for AI agents with hybrid vector + keyword search, container-isolated indexes, and 11 MCP tools. .NET, Docker-ready.

ai memory C# stdio
4
View

Cogmemai MCP

Scan found nothing
by hifriendbot

Persistent cognitive memory for Claude Code. Cloud-first with semantic search, AI-powered extraction, and project scoping. Zero local databases.

ai memory TypeScript stdio
claude mcp add cogmemai-mcp -- npx -y cogmemai
3
View

MCP Context Server

Scan found nothing
by alex-feel

An MCP server that provides persistent multimodal context storage for LLM agents.

ai memory Python stdio
claude mcp add mcp-context-server -- uvx mcp-context-server
3
View

Context7

Scan found nothing
by upstash

Up-to-date code documentation for LLMs and AI code editors.

ai memory TypeScript stdio
claude mcp add context7 -- npx -y @upstash/context7-mcp
49.3k 2 tools
View

FAQ

What are the best MCP servers for AI Memory & Context?

These are the 20 highest-signal AI Memory & Context servers from a directory of 2805, ranked by stars, maintenance recency and what our automated scan returned. Nobody here has used them: this is a sort, not a review. Shodh Memory is top of the list.

How do I install these MCP servers?

Each server page includes one-click install commands for Claude Code, Cursor, and VS Code. Click on any server above to see its install instructions.

Are these MCP servers safe to use?

No. 2,316 of the 2,805 servers here were shallow-cloned and put through five keyword searches in March 2026; 489 were never scanned at all. A green badge means those five searches matched nothing, which is not the same as safe — a grep cannot follow a variable or tell you what a server does once it is running. Read what the scan actually checks at /mcp/security, and read the source before you run anything.

Use the shortlist, then open a concrete workflow

Best-of pages are useful for narrowing options. The next practical move is a specific server detail page or a focused browser-first tool route.