Official MCP server for Memsolus — persistent memory for AI agents. Semantic search, knowledge graph, and auto-consolidation via Model Context Protocol.
概览
Memsolus is a persistent memory platform for AI agents. It lets your AI assistant remember context across conversations, track decisions, search past knowledge, and navigate entity relationships through a knowledge graph. This package exposes the full Memsolus API as an MCP (Model Context Protocol) server, so any MCP-compatible AI client — Claude Desktop, Claude Code, Cursor, Windsurf, VS Code Copilot, Zed, and others — can store and retrieve memories, query knowledge, and traverse entity graphs without custom integration code. - A Memsolus account (sign up) - An API key (see Getting an API Key below) - Node.js 18+ or Bun 1. Sign up or log in at app.memsolus.com 2. Go to > 3. Click 4. Give it a name (e.g., Claude Desktop) 5. Select permissions: at minimum MemoryRead and MemoryWrite. For full access, also enable KnowledgeRead and DashboardRead 6. Copy the key — it starts with msk_ and you will not be able to see it again after closing the dialog 7.
README
以下为原文 README 节选,完整内容请查看原文。
@memsolus/mcp
MCP Server for Memsolus — persistent memory, knowledge graphs, and semantic search for AI agents
Documentation | npm | GitHub | memsolus.com
What is Memsolus?
Memsolus is a persistent memory platform for AI agents. It lets your AI assistant remember context across conversations, track decisions, search past knowledge, and navigate entity relationships through a knowledge graph.
This package exposes the full Memsolus API as an MCP (Model Context Protocol) server, so any MCP-compatible AI client — Claude Desktop, Claude Code, Cursor, Windsurf, VS Code Copilot, Zed, and others — can store and retrieve memories, query knowledge, and traverse entity graphs without custom integration code.
Prerequisites
- A Memsolus account (sign up)
- An API key (see Getting an API Key below)
- Node.js 18+ or Bun
Getting an API Key
- Sign up or log in at app.memsolus.com
- Go to Settings > API Keys
- Click Create API Key
- Give it a name (e.g.,
Claude Desktop) - Select permissions: at minimum
MemoryReadandMemoryWrite. For full access, also enableKnowledgeReadandDashboardRead - Copy the key — it starts with
msk_and you will not be able to see it again after closing the dialog - Set the key as the
MEMSOLUS_API_KEYenvironment variable or paste it directly into your client config
Installation
Choose the section that matches your AI client. All configurations use npx -y @memsolus/mcp, which runs the latest published version without a local install step.
Claude Code
Run this command in your terminal:
claude mcp add memsolus -- npx -y @memsolus/mcp
To pass the API key inline without setting an environment variable:
claude mcp add memsolus -e MEMSOLUS_API_KEY=msk_your_key_here -- npx -y @memsolus/mcp
After adding, restart Claude Code or run /mcp to confirm that memsolus appears in the server list.
Claude Desktop
Locate and open your Claude Desktop configuration file:
- macOS:
~/Library/Application Support/Claude/claude_desktop_config.json - Windows:
%APPDATA%\Claude\claude_desktop_config.json
Add the following entry inside mcpServers:
{
"mcpServers": {
"memsolus": {
"command": "npx",
"args": ["-y", "@memsolus/mcp"],
"env": {
"MEMSOLUS_API_KEY": "msk_your_key_here"
}
}
}
}
安装
npx -y @memsolus/mcp配置
{
"mcpServers": {
"memsolus": {
"command": "npx",
"args": ["-y", "@memsolus/mcp"],
"env": {
"MEMSOLUS_API_KEY": "msk_your_key_here"
}
}
}
}