Wisdomgraph

io.github.cklam12345Published by GitHub account cklam12345v0.2.0Updated Oct 24, 2026

Neo4j-native DIKW memory for Claude Code — knowledge that compounds across sessions

VerifiedSTDIODesktop onlyDatabasesKnowledge & Memory

Overview

AI-generated overview

A Neo4j-backed memory server for Claude Code that stores ingested files, URLs and facts as a persistent, compounding knowledge graph.

What you can do
Ingest files, directories or URLs into the knowledge graph as memory nodes and store facts, decisions and insights explicitly as remembered entries. Run read-only Cypher traversals to query the stored graph. Promote nodes through a DIKW (data, information, knowledge, wisdom) pipeline and report tier counts with top Wisdom nodes. The graph is described as never resetting, so knowledge accumulates across sessions.
When to use it
Useful when you want Claude Code to keep durable, queryable memory across sessions and projects, especially if you already run Neo4j and want ingested documents and explicit notes organized into a knowledge hierarchy rather than a flat store.
Requirements
A local Neo4j instance (the README suggests a local DozerDB via a docker command) and the Python package installed with its MCP extra; the npm package is only a shim. A Neo4j connection (bolt URL and user) plus the WISDOM_NEO4J_PASSWORD secret, set via the connect command or exported directly. Desktop-only, stdio transport.
Before you install
Requires a Neo4j password (WISDOM_NEO4J_PASSWORD) that is stored in configuration; protect it and avoid committing it. Ingesting files, directories or URLs sends their contents into the graph, so avoid pointing it at sensitive material. Remembered facts and ingested data persist and are not reset, so mistakes stay in the graph.

Installation

In SourceWeft

  1. Open Wisdomgraph in the dashboard and add it to a workspace.
  2. Enable the server for the chats that should use its tools.

Desktop only via STDIO. STDIO servers start a local process, so they need the SourceWeft desktop host.

Other MCP clients

Follow the launch instructions in the repository.

README

wisdomGraph

Accumulative Neo4j-native DIKW memory for Claude Code — permanent knowledge across sessions.

[PyPI] [npm] [License: MIT]

This is the npm shim for wisdomGraph. All logic lives in the Python package.

Quick start

bash
# 1. Install Python packagepip install 'wisdomgraph[mcp]'
# 2. Start Neo4j (local DozerDB)wisdom docker upwisdom connect bolt://localhost:7687 --user neo4j --password password
# 3. Register MCP server in Claude Codewisdom mcp-install

MCP config (via npx)

json
{  "mcpServers": {    "wisdomGraph": {      "command": "npx",      "args": ["wisdomgraph"]    }  }}

What it does

wisdomGraph gives Claude Code permanent, accumulative memory backed by Neo4j.

Five MCP tools:

  • wisdom_ingest — absorb files, directories, or URLs
  • wisdom_remember — store facts, decisions, insights explicitly
  • wisdom_query — read-only Cypher traversal
  • wisdom_reflect — DIKW promotion pipeline
  • wisdom_report — tier counts + top Wisdom nodes

The graph never resets. Every session compounds.

Links

Source: npm/README.md at commit 1aab27b

Tools

0
SourceWeft has not started this package in its sandbox yet, so its tools are not known.

Version history

1
  1. v0.2.0LatestSep 16, 2026

Score

50% of the applicable checks met

How complete this listing is, computed from the registry entry, the endpoint and public sources. Not a rating of the server and not a statement that it is safe or working today: reachability is a dated check, and no installs or reviews go into the score.

  • Reached in the last 30 days25 pts

    Keep the endpoint answering the MCP handshake; SourceWeft probes it every 30 days.

  • Auth declared10 pts

    Declare the credential headers in the registry entry, or keep the endpoint answering the probe so its auth can be classified.

  • Tools listed15 pts

    Answer tools/list without a credential, or ship a package that starts in a sandbox without one.

  • DescriptionRequired15 pts
  • Logo10 pts
  • Documentation15 pts
  • License10 pts