skills.rest
michaeldowling
michaeldowlingCOMMUNITY
Explore all skills by this author →
Website𝕏@michaeldowlingGitHub

openclaw-memory-graph

Store and retrieve typed memory nodes as a queryable graph in Neo4j.

Updated Mar 28, 2026
Search Tags:#migration#embedding#neo4j#graph-memory#openclaw-memory-graph#vector-recall
One-click install
npx skills add https://github.com/michaeldowling/openclaw-memory-graph --skill openclaw-memory-graph
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: openclaw-memory-graph
Source: https://github.com/michaeldowling/openclaw-memory-graph/tree/main/skills/openclaw_memory_graph
Command: npx skills add https://github.com/michaeldowling/openclaw-memory-graph --skill openclaw-memory-graph

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Store and organize memory items as a typed, queryable graph in Neo4j, replacing brittle flat-file storage with semantic links and fast recall.

Core Features & Use Cases

  • Typed memory nodes (ShortTerm, LongTerm, WorkingMemory) with TTL support and migration of legacy memories.
  • Semantic edges and vector-based recall via embeddings, with provider options for OpenAI, Ollama, or local sentence-transformers.
  • Operations to save, recall, update, forget, connect, traverse, list, and prune memories across an agent.

Quick Start

Instantiate MemorySkill with an agent_id, then save a memory and recall memories relevant to a query.

Frequently Asked Questions about openclaw-memory-graph

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

FAQPage Schema
How do I store and recall agent memories as a queryable graph in Neo4j?▼

To store and recall agent memories as a graph in Neo4j, you can instantiate a memory skill with an agent_id to save typed nodes and perform vector recall using semantic edges and embeddings.

Can I migrate flat-file memories into a Neo4j graph database?▼

You can migrate flat-file memories into a Neo4j graph database using built-in migration operations that transform legacy memory items into typed nodes with semantic links.

What embedding providers can I use for vector recall with a Neo4j memory graph?▼

For vector recall with a Neo4j memory graph, you can use OpenAI, Ollama, or local sentence-transformers as embedding providers by configuring the respective environment variables.

Do I need a specific Neo4j version to manage typed memory nodes with TTL support?▼

Managing typed memory nodes with TTL support requires a running Neo4j 5.x instance and a Python environment configured with NEO4J_URI, NEO4J_USER, and NEO4J_PASSWORD environment variables.

How do short-term, long-term, and working memory lifecycles work in a graph database?▼

Memory lifecycles in a graph database work by categorizing nodes into ShortTerm, LongTerm, and WorkingMemory types, allowing operations to save, update, forget, and prune memories across an agent.

Related Skills

Explore More Software Engineering Skills→
MacPhobosMacPhobos

Better Auth Core

Configure Better Auth instances with environment variables, data drivers, and server handlers.

Community
Intermediate
ourmemourmem
198

memory-recall

Search the ourmem persistent memory store with natural language queries.

Official
Basic
zilliztechzilliztech
3

embedding

Encode text and images into vector representations for retrieval tasks.

Official
Intermediate
SyntaxAsSpiralSyntaxAsSpiral
5

memory-systems

Persist AI agent context across sessions with vector stores and temporal knowledge graphs.

Community
Advanced