MCP servers that ship as code — a repository or a package you install and run locally (usually over stdio). They have no hosted endpoint, so we cannot probe or grade them: instead each one shows its repository, install command, license, stars and when it was last updated.
Collected from GitHub repositories and npm, PyPI, Cargo, Go, Maven, NuGet and RubyGems packages tagged as MCP servers, the official MCP Registry, Docker's MCP catalog, ModelScope, Baidu MCP World, Cline and other public directories; one entry per repository. A server that also offers a hosted endpoint is listed under Remote instead. These entries are not counted in our reachability, verification or readiness statistics.
-
Shared document memory for AI-assisted writing. Reuse free templates, writing rules, and references for documents and slides through a web app, CLI, and MCP.
-
Self-hosted MCP server for shared AI project memory and human-approved living documentation across ChatGPT, Claude, and Codex.
npm Apache-2.0 -
内网 AI 知识库引擎(忆桥)· 企业文档问答 / 制度与合同检索 / 私有化部署 · 每个答案带逐句出处与可回放 traceId,版本链 + 审计 + 部门隔离 · Evidence-cited enterprise knowledge base & long-term memory MCP server for internal documents, local-first on SQLite + Ollama.
oci Apache-2.0 -
Starts the MCP server of the Tisty app installed on this computer: local memory and context for you and your AI agent, with notes, documents, tasks and ideas as Markdown on your own disk.
npm Apache-2.0 -
ISO 9001 7.5.3 / ISO 13485 4.2.5 document-control memory for AI agents. Nothing is deleted. EMET 0.3.6 — Enduring Memory & Epistemic Tiers. An MCP memory server with persistent shared memory for AI agents, bot teams and chats: layered memory, verified wri
npm MIT -
textview-mcpgithub.com/mrliuzhiyu/textview-mcpupdated this year
MCP server for TextView — persistent memory for AI agents. Save, search, and retrieve documents across AI sessions.
npm MIT -
EMET is an MCP memory server built for ISO 9001 7.5.3 / ISO 13485 4.2.5 document control: shared memory for AI agents and bot teams. Nothing is deleted; records are superseded, retired and retained. Controlled templates, verified writes, a word-for-word transcript tier, a GEM socket for add-on modules, stdio and remote HTTP with OAuth, on MongoDB.
npm MIT ★ 1 -
Shared long-term memory, document search and a code index for AI agents, linked to your code. Memories point at the files and symbols they're about and surface when an agent edits them. MCP server, Claude Code and Codex plugins, SDKs. Self-hosted on ClickHouse.
pypi Apache-2.0 ★ 114 -
mcp-docs-memorygithub.com/kroq86/mcp-docs-memoryupdated this year
MCP server for semantic document search and durable memory using Ollama embeddings and DuckDB.
-
Portable, governed, model-agnostic memory for AI agents: facts invalidated never deleted, raw text as truth, one documented export format.
npm Apache-2.0 ★ 3 -
A protocol-driven RAG system that decouples LLM reasoning from tool execution using MCP, enabling scalable document intelligence with ChromaDB and OpenAI. Implements an MCP client–server architecture with explicit message orchestration, structured tool execution, and stateful multi-turn memory for document-grounded LLM applications
-
flowsim-tutorgithub.com/Ahmed-Hassan-portfolio/flowsim-tutorupdated this year
Hybrid RAG + MCP documentation tutor with Chainlit UI, workflow memory, synthetic docs, and multi-server orchestration.
MIT -
Fully open-source memory for AI agents: cited recall, user Profiles, document Resources and native multi-node deployment on PostgreSQL.
npm Apache-2.0 ★ 2 -
Fact, not memory. Up-to-date library documentation for AI agents over MCP.
npm MIT -
Long-term memory for AI agents — an MCP server over Postgres + pgvector. Semantic recall across thoughts and documents, self-deduplicating capture, and namespace isolation by credential.
MIT ★ 1 -
Shared memory and documentation search for AI agents over MCP, on Cloudflare Workers.
-
PowerPoint MCP server — Claude & Cursor PowerPoint generator with zero-retention in-memory processing for confidential documents
-
Documentation RAG for coding agents: vector search, knowledge graph, agent memory, and an MCP server — torch-free local embeddings via fastembed
pypi Apache-2.0 ★ 1 -
Local memory + semantic search for AI agents. Indexes code and documents on your machine, queryable via MCP. Private, offline, fast. macOS arm64 + Linux x86_64.
★ 8 -
graphiti-foldergithub.com/notx2wice/graphiti-folderupdated this year
Append-only Graphiti memory and read-only MCP for a folder of Markdown and text documents.
MIT -
Private, evidence-first document agent with persistent local memory, exact citations, reversible actions, and bounded Nemotron reasoning.
MIT ★ 1 -
longform-memory-mcpgithub.com/emberspun/longform-memory-mcpupdated this year
MCP server giving your AI writing assistant a constant-size memory of a long document. Chapter 1000 costs the same tokens as chapter 10. Local-only: no account, no API key, no network.
npm MIT ★ 1 -
longform-memory-mcpgithub.com/emberspun/longform-memory-mcpupdated this year
MCP server that gives your AI writing assistant a constant-size memory of a long document. Chapter 1000 costs the same tokens as chapter 10. Local-only: no account, no API key, no network.
npm MIT -
Collective AI memory for multi-agent teams. 10-stage GAMR search engine, knowledge graph (Apache AGE), document ingestion, 40 MCP tools.
NOASSERTION ★ 11 -
MCP server for Ossian — ask your own documents with citations, and give an agent memory that persists
npm Apache-2.0 ★ 2 -
Local-first memory, hybrid RAG, and agent personalization for AI coding agents (OpenCode, Claude Code, Codex, Gemini CLI, Antigravity). MCP server + CLI with persistent context, document ingestion, vector + SQLite FTS5 retrieval, sync, setup, and safe uni
npm MIT -
KR Mycelia Memory – A native Windows 11 AI memory system powered by a Dynamic Associative Database (DAD). Persistent local memory, offline-first AI integration, MCP agent access, semantic associations, document and media processing, and transparent source-backed recall. Built in C++20.
NOASSERTION -
Evidence-first local context engine for AI agents - durable project memory for code, documents, and model-adaptive ContextPacks.
NOASSERTION -
Embedded NoSQL Database for TypeScript & AI Agents. Zero-native document database with ACID transactions & in-memory cache — verified on Node.js & Bun.
npm MIT -
Shared, persistent memory for any AI agent. Agents recall what your workspace established — instructions, decisions, facts, preferences, active working threads, and grounded PDF/DOCX/CSV/web documents — before they act, then report outcomes so confidence
npm Apache-2.0 -
Turn your documents into agent memory. MCP server with hybrid semantic + keyword search over emails, PDFs, notes, and past conversations. Works with Claude, ChatGPT, Cursor, and any MCP client.
MIT ★ 6 -
claude-crowedgithub.com/TheNewJavaman/claude-crowedupdated this year
A persistent semantic memory system for Claude Code, built as an MCP server. Replaces Claude Code's built-in flat-file memories with a structured, searchable, versioned document store.
Apache-2.0 -
MCP server for HotKV: lets AI agents read and write keys, agent memory, prompts and RAG documents.
Apache-2.0 -
Local-first, offline-capable Agent memory & knowledge service. REST + MCP dual-protocol, hybrid retrieval (vector + BM25/RRF), document/webpage RAG. No LLM required for storage & retrieval
pypi Apache-2.0 ★ 5 -
Local-first decision memory for AI coding agents. Rust library, CLI and MCP server for retrieving rationale from Git, ADRs and documents with source evidence and history.
Apache-2.0 ★ 1 -
DSH for Teams — Deploy DSH once on your LAN; memory, knowledge base & document search included. Team-ready, up to ~20 users. MIT-licensed.
MIT ★ 11 -
cgh, the context graph hub: the project memory of your coding agents, plus a code and document graph. Downloads and runs the standalone cgh binary; no Python required.
npm Apache-2.0 -
Self-hosted memory for your LLM conversations: import your chat archives and documents, search them, and chat over them on your own machine with Ollama and share that memory with any MCP client.
Apache-2.0 -
Portable memory for AI agents — the open MCP client for SAIHM. Eight tools any MCP agent uses to reach a SAIHM operator endpoint: encrypted, shareable, provably erased. Implements the publicly documented SAIHM memory protocol (Independent-Submission Internet-Draft; not an IETF standard). Evaluate it offline first with the runnable demos.
npm Apache-2.0 ★ 4 -
MCP plugins for Claude Code, Cursor, Gemini CLI and VS Code that give your AI agent local memory and context in Tisty: tasks, notes and documents on your disk.
npm Apache-2.0 ★ 1 -
Local memory and context for you and your AI agent: Markdown notes, documents and tasks on your disk, with an MCP server for GitHub Copilot, Claude Code, Codex and Cursor. Open source, no account.
AGPL-3.0 ★ 10 -
Local memory infrastructure for AI agents. Typed vaults (semantic, procedural and episodic) with document indexing, skill management, bi-temporal facts, and autonomous agent access via MCP, HTTP API, or CLI. Multi-vault architecture, zero cloud dependencies.
pypi MIT -
cghgithub.com/altikva/cghactive
cgh, the context graph hub: the project memory of your coding agents. Shared memory across sessions, agents and repos, plus a code and document graph that answers with exact file:line. 62 MCP tools for Claude Code, Codex, Cursor, Gemini and IBM Bob.
npm Apache-2.0 -
Identity, continuity & trust for AI agents — prove you're an LLM (not human), hold a spirit key for signing/encryption/memory, and issue offline-verifiable signed documents.
AGPL-3.0 -
MemorizedMCP delivers a high‑performance, local‑first MCP Server in Rust that equips AI agents with an advanced hybrid memory system. It combines a knowledge graph, vector embeddings, and full‑text search with a novel documentary memory capability, enabling accurate recall of exact documents alongside rich semantic relationships.
MIT ★ 8 -
Memory integrity for DataHub agents — agent memories that expire when the schema, glossary, or ownership beneath them changes. Built on DataHub's Timeline API, Documents and structured properties, with an MCP server for agents.
Apache-2.0 -
atlas-agent-intelligence-layergithub.com/adrianofratelli-glitch/atlas-agent-intelligence-layeractive
AI agent intelligence layer in MongoDB Atlas: prompts, model config, semantic cache, guardrails and short/long-term memory stored as documents and changed live, plus an MCP tool-use loop.
MIT -
orvugithub.com/rykth/orvuactive
Shared memory for teams of AI agents. Notes, tasks, budgets, documents and an inbox that many agents read and write at once, on one machine or many, merging automatically with no server.
Apache-2.0 ★ 1 -
sensei-mcpgithub.com/amarodeabreu/sensei-mcpupdated this year
Your engineering standards mentor - Active context injection for 50+ file types with session memory. Sensei transforms your engineering standards from passive documentation into an active mentor that injects relevant guidelines **before** Claude reasons, maintaining session memory of architectural decisions.
pypi Apache-2.0 ★ 1 -
ainative-prd-mcpgithub.com/AINative-Studio/ainative-prd-mcpupdated this year
AINative PRD Generator MCP Server — Generate, validate, and manage Product Requirement Documents with full AINative platform awareness and ZeroDB-powered persistent memory
npm MIT -
Local hybrid search for agents and humans. 20+ document formats, agent memory, zero cloud dependencies. MCP server + CLI.
pypi MIT ★ 7 -
ContextOS — document retrieval and conversation memory for AI workflows. MCP server, FastAPI, Google embeddings, and Pinecone. Keep the thread.
MIT ★ 4 -
everspring-mcpgithub.com/alpeerkaraca/everspring-mcpupdated this year
everspring-mcp is a production-oriented Model Context Protocol (MCP) server built to provide AI agents with accurate, version-specific Spring ecosystem documentation. Its primary objective is to reduce hallucination and version drift by grounding responses in a curated, indexed documentation base rather than relying on model memory.
★ 1 -
Shared context and memory coordination layer for swarms of parallel AI agents. Documented HTTP + WebSocket protocol, interoperable Python and Node CLIs, BM25 keyword memory search, built-in MCP server for agent tool calls.
npm MIT -
LiteLLM proxy hook and MCP server for chat bots in groups: three-scope memory (person, group, shared) whose scope is signed by the gateway, never chosen by the model; automatic recall; document routing; and an output filter that turns a tool call written as text into an honest admission.
MIT -
Comprehensive MCP server exposing dozens of capabilities to AI agents: multi-provider LLM delegation, browser automation, document processing, vector ops, and cognitive memory systems
pypi NOASSERTION ★ 164 -
2ndbraingithub.com/mull2536/2ndbrainupdated this year
Local Windows-native MCP server giving Claude Code verbatim cross-session recall - SQLite FTS5 over session transcripts, memory files and documents. Zero external LLM calls, credential redaction at ingest.
-
🧠 Transform documentation chaos into a structured memory system with Mnemos, your self-hosted, multi-context knowledge server for developers.
MIT ★ 5 -
Read-only MCP server for semantic search over a local document corpus — on-device embeddings (Ollama), local Chroma store, symlink-hardened file access. Sibling of claude-code-session-memory.
MIT -
8 specialized MCP servers for AI-assisted development: code analysis, semantic refactoring, test running, documentation validation, critique-refine, code graph, memory, and GitHub integration.
MIT