# Knowledge Graph Memory

> Use this tool when you need to create a persistent memory system for language models, enabling them to store and retrieve complex knowledge structures over time. It solves problems such as remembering user preferences, tracking relationships, and accumulating domain knowledge across conversations. The Knowledge Graph Memory server takes in API queries and updates, and outputs stored entities, relations, and observations, making it ideal for use cases requiring personalized AI assistants.

Canonical page: https://skillsregistry.net/skills/modelcontextprotocol-memory  
JSON: https://api.skillsregistry.net/v1/skills/modelcontextprotocol-memory

## Description

This knowledge graph memory server, developed by Anthropic, provides a persistent memory system for language models using a local graph database. It enables AI agents to create, query, and update entities, relations, and observations through a controlled API. By storing information as a semantic network, it allows AI systems to build and maintain complex knowledge structures over time. The server integrates with Claude Desktop via NPM and focuses on personalization use cases. This implementation is particularly useful for AI assistants designed to remember user preferences, track relationships between people and organizations, or accumulate domain knowledge across conversations.

## Trust

- **Trust score (0–1):** 0.50
- **Verification tier:** unverified

## Facts

- **Version:** 1.0.0
- **Skill type:** atomic
- **Execution layer:** mcp-remote
- **Runtime environment:** api
- **Category:** database
- **Updated:** 2026-04-29

## Source

- **Source listing:** [PulseMCP](https://www.pulsemcp.com/servers/modelcontextprotocol-memory)
- **Repository:** <https://github.com/modelcontextprotocol/servers/tree/HEAD/src/memory>

## Use it

Resolve this record through the SkillsRegistry MCP server (no auth, read-only):

```
claude mcp add --transport http --scope user skillsregistry https://api.skillsregistry.net/mcp
```

```json
{
  "jsonrpc": "2.0",
  "id": 1,
  "method": "tools/call",
  "params": {
    "name": "get_skill",
    "arguments": {
      "slug": "modelcontextprotocol-memory"
    }
  }
}
```

REST: `GET https://api.skillsregistry.net/v1/skills/modelcontextprotocol-memory` · pull for local use: `GET https://api.skillsregistry.net/v1/skills/modelcontextprotocol-memory/pull`

---
SkillsRegistry indexes agent skills from public registries and GitHub. Skills we have analysed are scanned with Circle-IR and scored on six dimensions; each listing states its scan coverage. More: https://skillsregistry.net/llms.txt