Best MCP Servers for Memory & Knowledge (2026)

The best MCP servers for memory and knowledge — give agents persistent memory, note stores, and searchable knowledge bases that survive across sessions.

Top 10 · ranked by installs, upvotes, stars and views · refreshed hourly

  1. Why it’s hereMost popular in this category on AllMCPs

    MCP & Agent Skills for Automated Documentation, and codebase conventions + context

    Upvotes: 4Installs: 25GitHub stars: 2
    Install
  2. Why it’s hereOfficial server

    Dynamic and reflective problem-solving through thought sequences.

    Upvotes: 0Installs: 3GitHub stars: 91k
    Install
  3. Why it’s herePopular pick — 1 installs, 1 upvotes

    Code-intelligence engine that indexes a repo into a persistent knowledge graph — functions, classes, call chains, HTTP routes, cross-service links. 159 languages via tree-sitter + Hybrid LSP, sub-ms structural queries, 99% fewer tokens than grep. Single static binary, zero dependencies, 100% local. npx codebase-memory-mcp

    Upvotes: 1Installs: 1GitHub stars: 47k
    Install
  4. Hindsight logo

    #4 Hindsight

    vectorize-io

    Why it’s hereCommunity pick

    Hindsight: Agent Memory That Works Like Human Memory - Built for AI Agents to manage Long Term Memory

    Upvotes: 0Installs: 0GitHub stars: 54k
    Install
  5. Why it’s hereCommunity pick

    Universal AI bridge for Obsidian vaults using MCP. Provides safe read/write access to notes with 11 comprehensive methods for vault operations including search, batch operations, tag management, and frontmatter handling. Works with Claude, ChatGPT, and any MCP-compatible AI assistant.

    Upvotes: 0Installs: 0GitHub stars: 1.7k
    Install
  6. Why it’s hereCommunity pick

    MCP server for Obsidian vaults — read, write, search, and edit notes via the Local REST API plugin.

    Upvotes: 0Installs: 0GitHub stars: 701
    Install
  7. Deja Vu logo

    #7 Deja Vu

    vshulcz

    Why it’s hereCommunity pick

    Local memory layer over the session histories coding agents already write (Claude Code, Codex CLI, opencode): lexical search, recall tools, session-start auto-recall, secret redaction at index time, cross-machine sync over SSH.

    Upvotes: 0Installs: 0GitHub stars: 1.3k
    Install
  8. Why it’s hereCommunity pick

    The private, owned conversation-memory layer for AI. Record, transcribe, and search every meeting.

    Upvotes: 0Installs: 0GitHub stars: 1.6k
    Install
  9. Why it’s hereCommunity pick

    Privacy-first document search server running entirely locally. Supports semantic search over PDFs, DOCX, TXT, and Markdown files with LanceDB vector storage and local embeddings - no API keys or cloud services required.

    Upvotes: 0Installs: 0GitHub stars: 425
    Install
  10. Why it’s hereCommunity pick

    Local-first persistent project memory for AI coding agents across MCP clients and sessions.

    Upvotes: 0Installs: 0GitHub stars: 873
    Install

Which Memory & Knowledge MCP server should you use?

When choosing a Model Context Protocol server for Memory & Knowledge, select verified or official listings if you need strict API security guarantees. For rapid local dev testing with Claude or Cursor, community-maintained tools offer zero-setup configuration blocks ready to copy-paste.

Frequently asked questions about Memory & Knowledge MCP servers

What is the best MCP server for agent memory?

Memory servers range from simple key-value stores to vector-backed knowledge bases. This page ranks them by usage so you can pick a proven approach.

How does MCP give an AI agent long-term memory?

A memory MCP server persists information outside the model’s context window and exposes tools to store and retrieve it, so the agent can recall facts in later sessions.

Can I use my own notes or docs as agent knowledge?

Yes — several servers index local files, note apps, or knowledge bases so agents can search and cite your own content.