mcp-neo4j-memory-ops

Merge duplicate memory nodes and edges in Neo4j MCP.

1|Updated Jan 15, 2026
One-click install
npx skills add https://github.com/janjaszczak/cursor --skill mcp-neo4j-memory-ops
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: mcp-neo4j-memory-ops
Source: https://github.com/janjaszczak/cursor/tree/main/skills/mcp-neo4j-memory-ops
Command: npx skills add https://github.com/janjaszczak/cursor --skill mcp-neo4j-memory-ops

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill helps teams model and manage memories as graph data in Neo4j MCP, enabling consistent storage, de-duplication, and retrieval of related learnings across projects.

Core Features & Use Cases

  • Memory modeling: Create and relate MEMORY nodes (PROJECT, DECISION, CONSTRAINT, TOOL, WORKFLOW) with provenance data.
  • Deduplication & consistency: Merge duplicates and maintain a single source of truth for memories.
  • Retrieval & analysis: Traverse relationships to answer how memories relate to ongoing work and future decisions.

Quick Start

Connect to your Neo4j memory MCP, then run the memory-ops to create/update nodes and query relationships.

Frequently Asked Questions about mcp-neo4j-memory-ops

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I deduplicate and merge linked memories in a Neo4j graph?

To deduplicate linked memories in a Neo4j graph, you create or update nodes and edges to merge duplicates. This ensures a single source of truth by maintaining normalization and provenance across projects.

What is the best way to model project decisions and constraints as graph memory data?

Modeling project memory data involves creating nodes for PROJECT, DECISION, CONSTRAINT, TOOL, and WORKFLOW. You relate these nodes to track confidence and last_updated metadata for robust retrieval.

Can I retrieve connected relationships and provenance for global learnings using Neo4j MCP?

Yes, you can retrieve connected relationships and provenance for global learnings using Neo4j MCP. The workflow traverses graph edges to answer how memories relate to ongoing work and future decisions.

How do I query related memories across multiple projects in a graph database?

You query related memories across projects by applying a memory-ops workflow to traverse graph edges. This retrieves connected relationships and ensures normalization across different projects.

Does this memory modeling approach handle confidence tracking and error handling?

Yes, this memory modeling approach handles confidence tracking and error handling. It tracks confidence and last_updated metadata while applying robust error handling during node and edge creation.