Easy Memory
About
Easy Memory MCP - Dual-engine embedding service with Qdrant vector store
Details
- Author
- FlippySun
- GitHub stars
- 3
- Downloads
- 371
- Categories
- Knowledge Base, AI
Jump to
- Dual shell architecture: MCP (stdio) and HTTP (RESTful API)
- Remote proxy mode – no local Qdrant or Ollama required
- Supports multiple AI clients: Claude Desktop, Cursor, VS Code
- Persistent memory across sessions and projects
- Semantic search with hybrid retrieval (vector + BM25)
- 8 MCP tools: save, search, forget, status (with aliases)
Setting up with Highlight
This MCP is not yet compatible with Highlight’s one-click setup. However, you can still use it with Highlight by following these steps:
- Download and install Highlight from highlightai.com/download
- Navigate to the plugins tab and select "Add Custom Plugin"
-
Configure the plugin with the settings below
Plugin Name
Easy MemoryCommand (node, npx, python, etc.)Please refer to the README for specific instructions on how to obtain API keys or other required environment variables.
- Enable "Start Automatically" if you want the plugin to start when Highlight launches
From the repository
Easy Memory can be run via npx easy-memory, Docker Compose, or from source. Configure it by setting environment variables such as QDRANT_URL, QDRANT_API_KEY, and OLLAMA_BASE_URL for local setups, or EASY_MEMORY_URL and EASY_MEMORY_TOKEN for remote proxy mode. For client integration, edit the MCP configuration file (e.g., claude_desktop_config.json, .vscode/mcp.json) with the appropriate command and environment variables.
Claude Desktop / Cursor
Paste into your MCP client config file to install this server.
{
"mcpServers": {
"easy memory": {
"easy-memory": {
"command": "npx",
"args": [
"-y",
"easy-memory@latest"
],
"env": {
"EASY_MEMORY_URL": "https://memory.zhiz.chat/mcp",
"EASY_MEMORY_TOKEN": "<YOUR-TOKEN Get By: https://memory.zhiz.chat/>"
}
}
}
}
}McpServers
{
"easy-memory": {
"command": "npx",
"args": [
"-y",
"easy-memory@latest"
],
"env": {
"EASY_MEMORY_URL": "https://memory.zhiz.chat/mcp",
"EASY_MEMORY_TOKEN": "<YOUR-TOKEN Get By: https://memory.zhiz.chat/>"
}
}
}Easy Memory
> 让 AI 跨会话、跨项目持久化记忆。 基于 Qdrant 向量数据库 + 官方向量模型（支持经 OpenAI-compatible relay 接入）/ Ollama 的 MCP 记忆服务。
Easy Memory 提供双 Shell 架构：
- MCP Shell — 通过 stdio 与 Claude Desktop / Cursor / VS Code 等 IDE 直接通信
- HTTP Shell — RESTful API，支持远程访问、多客户端共享记忆
---
目录
- 快速开始
- 客户端配置
- 30 秒排障（远端优先）
- HTTP API
- MCP Tools
- Docker 部署
- 环境变量
- Web UI 管理面板
- 架构文档
---
快速开始
推荐路径（远端 VPS 优先，约 30 秒）
如果你已有远端 Easy Memory 服务（例如 https://memory.zhiz.chat）和 API Key（em_...），建议优先使用 stdio 远程代理模式：
1. 在客户端 MCP 配置里使用 npx easy-memory@latest + EASY_MEMORY_URL + EASY_MEMORY_TOKEN
2. 执行 Reload Window / 重启客户端
3. 运行 MCP: List Servers（或客户端等效命令）确认 server 已连接
> 根键速记：VS Code 的 .vscode/mcp.json 用 servers；Claude Desktop / Cursor 常见 mcpServers。
> 只有在你没有远端服务时，再走下方“本地自托管（Qdrant + Ollama）”路径。
前置条件
| 依赖 | 版本 | 说明 |
| ------- | ---- | ------------------------------------------ |
| Node.js | ≥ 20 | 运行 MCP Server |
| Docker | ≥ 24 | 运行 Qdrant + Ollama（或使用远程模式跳过） |
> 💡 远程代理模式无需本地 Qdrant/Ollama
> 如果你有远端 Easy Memory 服务（如 memory.zhiz.chat），只需设置 EASY_MEMORY_TOKEN 和 EASY_MEMORY_URL 即可。参见 远程代理模式。
基础服务准备（Qdrant + Ollama）
Easy Memory 依赖两个基础服务：Qdrant（向量数据库）和 Ollama（本地 Embedding 模型）。
```bash
Sign in to leave a review
Use Google, GitHub, or an email account so ratings stay tied to real people.
No reviews posted yet.
