MCP Neo4j Memory Server

@sylweriusz

代码仓库

基本信息

分类

记忆与知识

许可证

MIT

传输方式

stdio

作者

sylweriusz

关于 MCP Neo4j Memory Server

暂无概览

MCP Neo4j Memory Server 配置

使用下面的配置,将此服务器添加到你的 MCP 客户端。

{
  "mcpServers": {
    "mcp-neo4j-memory-server": {
      "command": "docker",
      "args": [
        "run",
        "\\"
      ]
    }
  }
}

MCP Neo4j Memory Server 工具

工具列表暂未提供

当前收录暂时没有可展示的工具列表，不代表服务没有工具。请查阅项目文档了解支持的能力。

工具是从 README 中自动提取的。维护者可以在 ## Tools 标题下列出工具,即可填充这部分内容。

概览

What is MCP Neo4j Memory Server?

An MCP server that gives AI assistants persistent, intelligent memory using Neo4j’s graph database. It stores memories as interconnected knowledge nodes, supports semantic and exact search, creates typed relationships, organizes data by project, and tracks knowledge evolution over time.

How to use MCP Neo4j Memory Server?

Install via npm install @sylweriusz/mcp-neo4j-memory-server, then add the tool to your Claude Desktop config with environment variables NEO4J_URI, NEO4J_USERNAME, and NEO4J_PASSWORD. You must run a Neo4j database with the GDS plugin (recommended: DozerDB). Four unified MCP tools—memory_store, memory_find, memory_modify, and database_switch—handle all memory operations.

Key features of MCP Neo4j Memory Server

Stores memories as nodes with typed relationships and temporal metadata
Unified search: vectors, exact matches, wildcards, and graph traversal
Creates multiple memories with relations in a single batch request
Isolates projects via separate databases with instant switching
Filters memories by relative or absolute date on any temporal field
Zero-fallback error handling for reliable debugging

Use cases of MCP Neo4j Memory Server

Remember facts, preferences, and context across AI conversations
Find relevant memories using semantic search or exact ID lookups
Create meaningful connections between memories with cross-references
Organize and isolate memories for different projects or topics
Track how knowledge changes over time with versioned observations

常见问题

What database does the server require?

It requires Neo4j with the Graph Data Science (GDS) plugin. The recommended setup is DozerDB (docker pull graphstack/dozerdb). GDS is necessary for vector operations.

What are the four MCP tools?

`memory_store` (create memories with relations), `memory_find` (search/retrieve), `memory_modify` (update, delete, manage observations/relations), and `database_switch` (change project context).

How do I troubleshoot vector search failures?

Check the server logs for `[VectorSearch] GDS Plugin detected`. If missing, ensure the GDS plugin is installed and Neo4j is running with the correct configuration.

Can I use this server without a graph database?

No. Neo4j (with GDS) is a required persistent storage layer. The server has no fallback or in‑memory mode.

How do I connect to an existing Neo4j instance?

Set the `NEO4J_URI`, `NEO4J_USERNAME`, and `NEO4J_PASSWORD` environment variables in your MCP client config. Test connectivity via `curl http://localhost:7474`.

记忆与知识分类下的更多 MCP 服务器

Memory

modelcontextprotocol

Model Context Protocol Servers

88.1K

PLUR

plur-ai

AI agents start every session with amnesia — you re-explain the project, repeat your preferences, and correct the same mistakes over and over. PLUR gives them a memory that persists. Your agent's corrections, preference

226

Agimem

Persistent memory for AI agents. Hosted MCP server with capsule-scoped key/value storage, tags, TTL expiration, bulk ops, and RBAC.

Scento Perfume API (MCP server): free fragrance API, no key

Scento

Scento Perfume API is a free fragrance API and perfume notes database for developers and AI agents. Search 1,500+ perfumes and 400+ ingredients, compare scents, find similar perfumes, map scent families and build occasio

M1K3

Round-Tower

Local AI agent for your Mac with a built-in MCP server: 18 tools for your notes and PDFs, a temporal memory graph, an on-device LLM and a neural voice. No account, no telemetry. No data centers required.