Memory MCP Worker

Memory MCP Worker

Provides cross-device access to a persistent knowledge graph via Cloudflare Workers, enabling memory storage and retrieval through both MCP protocol and REST API with full-text search capabilities.

Category
Visit Server

README

Memory MCP Worker

Cloudflare Worker implementing MCP protocol for cross-device memory/knowledge graph access.

Features

  • MCP Protocol Support: Works with Claude Code via HTTP transport
  • REST API: Access from mobile apps, desktop apps, n8n workflows
  • D1 Database: SQLite-based persistent storage
  • Full-text Search: Search across all observations

Quick Start

1. Install Dependencies

npm install

2. Create D1 Database

# Create the database
wrangler d1 create memory-graph

# Copy the database_id from output to wrangler.toml

3. Run Migrations

# Local development
npm run db:migrate:local

# Production
npm run db:migrate

4. Deploy

npm run deploy

5. Set API Key Secret

# Generate an API key
openssl rand -base64 32 | tr -d '/+=' | head -c 32

# Add as secret
echo "YOUR_API_KEY" | npx wrangler secret put API_KEY

6. Configure Claude Code

Add to your ~/.claude/settings.json or .mcp.json:

{
  "mcpServers": {
    "memory-remote": {
      "type": "http",
      "url": "https://memory-mcp.45black-limited.workers.dev/mcp",
      "headers": {
        "X-API-Key": "YOUR_API_KEY"
      }
    }
  }
}

Migrating from Local Memory

  1. Export your local memory:

    claude -p "Use mcp__memory__read_graph and output the raw JSON result" > scripts/local-memory.json
    
  2. Run migration:

    npx ts-node scripts/migrate-local-memory.ts
    

API Endpoints

REST API (for apps)

Endpoint Method Description
/api/entities GET List all entities
/api/entities/:name GET Get entity with observations
/api/entities POST Create entity
/api/entities/:name/observations POST Add observations
/api/relations GET List all relations
/api/relations POST Create relation
/api/search?q= GET Search entities and observations
/api/graph GET Get full graph
/api/import POST Bulk import

MCP Protocol

Endpoint Method Description
/mcp POST MCP JSON-RPC requests
/mcp/sse GET SSE connection for MCP

MCP Tools Available

  • create_entities - Create new entities
  • create_relations - Create relations between entities
  • add_observations - Add observations to entities
  • read_graph - Read entire knowledge graph
  • search_nodes - Search by query
  • open_nodes - Get specific nodes by name
  • delete_entities - Delete entities
  • delete_relations - Delete relations

Authentication

All /api/* and /mcp/* endpoints require authentication via API key.

Headers supported:

  • X-API-Key: YOUR_API_KEY
  • Authorization: Bearer YOUR_API_KEY

The health check endpoint (/) is public.

Mobile App Integration

Example fetch from a mobile app:

const API_KEY = 'YOUR_API_KEY';
const BASE_URL = 'https://memory-mcp.45black-limited.workers.dev';

const headers = {
  'Content-Type': 'application/json',
  'X-API-Key': API_KEY,
};

// Search for a project
const response = await fetch(`${BASE_URL}/api/search?q=Household%20Planner`, { headers });
const data = await response.json();

// Add an observation
await fetch(`${BASE_URL}/api/entities/Household%20Planner/observations`, {
  method: 'POST',
  headers,
  body: JSON.stringify({
    contents: ['Added new budget category feature']
  })
});

n8n Integration

An n8n workflow is included for querying the Memory API via webhooks.

Setup

  1. Import n8n-workflow-memory-api.json into n8n
  2. Create a "Header Auth" credential:
    • Name: Memory MCP API Key
    • Header Name: X-API-Key
    • Header Value: YOUR_API_KEY
  3. Assign the credential to all three HTTP Request nodes
  4. Activate the workflow

Endpoints

Search - POST to /webhook/memory-search:

curl -X POST https://your-n8n-instance/webhook/memory-search \
  -H "Content-Type: application/json" \
  -d '{"query": "Household Planner"}'

Get Entity - POST to /webhook/memory-entity:

curl -X POST https://your-n8n-instance/webhook/memory-entity \
  -H "Content-Type: application/json" \
  -d '{"name": "Household Planner"}'

Get Full Graph - POST to /webhook/memory-graph:

curl -X POST https://your-n8n-instance/webhook/memory-graph \
  -H "Content-Type: application/json" \
  -d '{}'

Create Entity - POST to /webhook/memory-create:

curl -X POST https://your-n8n-instance/webhook/memory-create \
  -H "Content-Type: application/json" \
  -d '{"name": "My Project", "entityType": "Project", "observations": ["First observation"]}'

Add Observations - POST to /webhook/memory-observations:

curl -X POST https://your-n8n-instance/webhook/memory-observations \
  -H "Content-Type: application/json" \
  -d '{"name": "My Project", "contents": ["New observation 1", "New observation 2"]}'

Create Relation - POST to /webhook/memory-relations:

curl -X POST https://your-n8n-instance/webhook/memory-relations \
  -H "Content-Type: application/json" \
  -d '{"from": "Project A", "to": "Project B", "relationType": "depends_on"}'

Delete Entity - POST to /webhook/memory-delete:

curl -X POST https://your-n8n-instance/webhook/memory-delete \
  -H "Content-Type: application/json" \
  -d '{"name": "Entity Name"}'

Delete Relation - POST to /webhook/memory-delete-relation:

curl -X POST https://your-n8n-instance/webhook/memory-delete-relation \
  -H "Content-Type: application/json" \
  -d '{"from": "Entity A", "to": "Entity B", "relationType": "relation_type"}'

Development

# Start local dev server
npm run dev

# Typecheck
npm run typecheck

Architecture

┌─────────────────────────────────────────┐
│         Cloudflare Worker               │
│      memory-mcp.45black.workers.dev     │
├─────────────────────────────────────────┤
│  Hono Framework                         │
│  ├── /api/* → REST API                  │
│  └── /mcp   → MCP Protocol              │
├─────────────────────────────────────────┤
│  Cloudflare D1 (SQLite)                 │
│  ├── entities                           │
│  ├── observations                       │
│  ├── relations                          │
│  └── observations_fts (full-text)       │
└─────────────────────────────────────────┘

Recommended Servers

playwright-mcp

playwright-mcp

A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.

Official
Featured
TypeScript
Audiense Insights MCP Server

Audiense Insights MCP Server

Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.

Official
Featured
Local
TypeScript
Magic Component Platform (MCP)

Magic Component Platform (MCP)

An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.

Official
Featured
Local
TypeScript
VeyraX MCP

VeyraX MCP

Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.

Official
Featured
Local
Kagi MCP Server

Kagi MCP Server

An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.

Official
Featured
Python
graphlit-mcp-server

graphlit-mcp-server

The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.

Official
Featured
TypeScript
Qdrant Server

Qdrant Server

This repository is an example of how to create a MCP server for Qdrant, a vector search engine.

Official
Featured
Neon Database

Neon Database

MCP server for interacting with Neon Management API and databases

Official
Featured
Exa Search

Exa Search

A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.

Official
Featured
E2B

E2B

Using MCP to run code via e2b.

Official
Featured