Mcp-server-neo4j-memory

Neo4j Memory is a Model Context Protocol (MCP) server that gives AI assistants long-term memory by storing past conversations and facts in a Neo4j graph…

  • AI
  • Databases
Neo4j Memory is a Model Context Protocol (MCP) server that gives AI assistants long-term memory by storing past conversations and facts in a Neo4j graph database. It organises information as a connected web of nodes so responses become more accurate, personalised, and context-aware over time.

Platform availability

Available on 0 of 10 platforms

  • Unraid (not listed)
  • TrueNAS (not listed)
  • Umbrel (not listed)
  • ZimaOS (not listed)
  • Proxmox (not listed)
  • Helm (not listed)
  • Mac (not listed)
  • Windows (not listed)
  • Linux (not listed)
  • Any server (not listed)

Health score?A 0-100 score computed nightly from four factors: maintained (40%: recent commits and steady activity), popular (25%: GitHub stars, log scale), easy to install (20%: platforms, install notes, setup guides, Docker image) and light to run (15%: minimum RAM, ARM support). When the source code has been checked, a fifth factor, security check (20%: automated scan for committed secrets, unsafe desktop-app settings, known-vulnerable dependencies and project hygiene; any serious finding caps it at 20), is added. Factors without data are left out and the rest are rescaled.

58/100 Fair

Maintained
68last commit 73 days ago
Popular
60991 GitHub stars
Easy to install
250 install platforms, install notes, Docker image
Light to run
No data
Security check
70no serious findings, 5 minor, CI, tests

Based on 4 of 5 factors.

GitHub stars
991
Open issues
32
Last commit
2026-09-09
Latest release
mcp-neo4j-cypher-v0.6.0
License
MIT
Activity
stale

Checked 3 days ago - source: GitHub (neo4j-contrib/mcp-neo4j)

Resources & Compatibility

ARM (e.g. Raspberry Pi) is supported.

Checked 54 days ago - source: Docker Hub image tags and published docs

First-install notes

  • Install the memory commands first Install the package globally with npm so the memory and mcp-local-memory commands are available. If you skip this, commands like memory --help will not be found on your system. source
  • Start the server with port and token Start the SSE (a streaming web connection) server on port 8320 (the number it listens on) with a secret token, for example with --transport sse --port 8320 --token 'your-secret-token' or docker compose up -d. If your client uses a different port or token, it will not reach the server or be allowed in. source
  • Add the server URL and token to client In your AI app's MCP (a tool protocol for AI apps) settings, use the server address ending in /sse and set the Authorization header to Bearer your-secret-token. If the token is missing or wrong, the server rejects the connection and the memory tools won't work. source

Fetched 53 days ago - each note links to its source

Alternatives (5)

Checked 40 days ago - source: AlternativeTo

Common questions

Can the same memory be used across different conversations, sessions, or clients?

Yes. It stores and retrieves entities and relationships from a personal knowledge graph in Neo4j, and the information can be accessed across different sessions, conversations, and clients.

Does it require a remote or cloud-hosted Neo4j instance?

No. It can use a local or remote Neo4j instance.

What memory operations does the server expose to the AI?

It provides memory_store, which creates memories with observations and immediate relations in one operation, and memory_find, which supports semantic search, direct ID lookup, date filtering, and graph traversal.

How does the server query the stored memory data in Neo4j?

It uses Cypher queries to explore and retrieve stored information and analyze relationships between data points.

Answers sourced from github.com, github.com, pypi.org

Community

Collected 48 days ago - public community threads