npm package discovery and stats viewer.

Discover Tips

  • General search

    [free text search, go nuts!]

  • Package details

    pkg:[package-name]

  • User packages

    @[username]

Sponsor

Optimize Toolset

I’ve always been into building performant and accessible sites, but lately I’ve been taking it extremely seriously. So much so that I’ve been building a tool to help me optimize and monitor the sites that I build to make sure that I’m making an attempt to offer the best experience to those who visit them. If you’re into performant, accessible and SEO friendly sites, you might like it too! You can check it out at Optimize Toolset.

About

Hi, 👋, I’m Ryan Hefner  and I built this site for me, and you! The goal of this site was to provide an easy way for me to check the stats on my npm packages, both for prioritizing issues and updates, and to give me a little kick in the pants to keep up on stuff.

As I was building it, I realized that I was actually using the tool to build the tool, and figured I might as well put this out there and hopefully others will find it to be a fast and useful way to search and browse npm packages as I have.

If you’re interested in other things I’m working on, follow me on Twitter or check out the open source projects I’ve been publishing on GitHub.

I am also working on a Twitter bot for this site to tweet the most popular, newest, random packages from npm. Please follow that account now and it will start sending out packages soon–ish.

Open Software & Tools

This site wouldn’t be possible without the immense generosity and tireless efforts from the people who make contributions to the world and share their work via open source initiatives. Thank you 🙏

© 2026 – Pkg Stats / Ryan Hefner

@lakphy/talk-to-me

v0.0.1

Published

A MCP Server that turns any Agent into a thoughtful personal assistant with memory.

Readme

Talk to Me

一个 MCP (Model Context Protocol) Server,让任何 Agent 都能成为你的贴心私人助理。

设计理念

  • 极简主义: 使用 LibSQL (SQLite) 本地存储,无需复杂的向量数据库
  • 原声保真: 严格记录用户原话,不做任何润色或总结
  • 历史留痕: 软删除机制,保留完整的思维演变轨迹
  • 智能回响: 写入时自动关联历史记录,防止信息孤岛
  • 语义交给 Agent: 不依赖字符串检索,让 Agent 自己做语义理解

快速开始

安装

方式 1: 全局安装(推荐)

npm install -g @lakphy/talk-to-me
# 或
pnpm add -g @lakphy/talk-to-me

方式 2: 使用 npx(无需安装)

直接在 MCP 配置中使用 npx 调用。

详细安装指南: 查看 INSTALL.md

配置

在你的 MCP 客户端(Claude Desktop 或 Cursor)配置文件中添加:

使用全局安装:

{
  "mcpServers": {
    "talk-to-me": {
      "command": "talk-to-me"
    }
  }
}

使用 npx(无需安装):

{
  "mcpServers": {
    "talk-to-me": {
      "command": "npx",
      "args": ["-y", "@lakphy/talk-to-me"]
    }
  }
}

配置文件位置

  • Claude Desktop (macOS): ~/Library/Application Support/Claude/claude_desktop_config.json
  • Claude Desktop (Windows): %APPDATA%\Claude\claude_desktop_config.json
  • Cursor: 在设置中搜索 "MCP"

数据存储

数据库文件位置遵循 XDG 规范:

  • macOS: ~/Library/Application Support/talk-to-me/memory.db
  • Linux: ~/.local/share/talk-to-me/memory.db
  • Windows: %APPDATA%/talk-to-me/memory.db

MCP 接口

Tools

capture_fragment

捕获用户的想法、待办、目标等碎片信息。

{
  content: string;      // 用户原话(红线:必须是原话)
  type: 'idea' | 'todo' | 'flag' | 'insight' | 'moment';
  tags?: string[];      // 标签
  keywords?: string[];  // 关键词
  mood?: string;        // 情绪
  energy_level?: number; // 精力水平 1-10
}

remove_fragment

软删除碎片记录(保留历史)。

{
  id: string;      // 碎片 ID
  reason?: string; // 删除原因
}

search_notes

获取或搜索碎片记录。推荐让 Agent 自己做语义筛选

{
  keyword?: string;          // 搜索关键词(可选,不传返回全部)
  type?: 'idea' | 'todo' | 'flag' | 'insight' | 'moment'; // 按类型筛选
  include_deleted?: boolean; // 是否包含已删除记录,默认 true
  limit?: number;            // 返回数量限制,默认 100
}

推荐用法:不传 keyword,通过 type 筛选或直接获取全部,让 Agent 自己做语义理解:

// 获取所有 flag
{ "type": "flag" }

// 获取全部记录
{ }

Resources

talk-to-me://context/now

获取当前上下文快照,包含:

  • 最近的碎片记录
  • 活跃的待办事项
  • 活跃的目标 (Flag)

talk-to-me://fragments/all

获取全部碎片记录(包含已删除),让 Agent 自己做语义理解和筛选。

Prompts

thoughtful_companion

将 Agent 转变为贴心的私人助理,包含完整的系统指令:

  • 原话记录红线
  • 时间精确参考
  • 软删除处理规则
  • 智能回响使用指南

使用示例

场景 1: 记录想法

用户: "我想健身"

Agent 调用 capture_fragment:

{
  "content": "我想健身",
  "type": "flag",
  "keywords": ["健身", "运动", "锻炼"]
}

场景 2: 查询 Flag

用户: "我立过什么 flag?"

Agent 调用 search_notes:

{
  "type": "flag"
}

返回全部 flag 类型记录,Agent 自己做语义理解和组织回复。

场景 3: 放弃目标

用户: "健身太累了,算了"

Agent 调用 remove_fragment:

{
  "id": "frag_xxx",
  "reason": "太累了"
}

场景 4: 回顾历史

用户: "我以前想做什么来着?"

Agent 调用 search_notes:

{
  "include_deleted": true
}

Agent 拿到全部记录后,自己分析并回复: "你之前想健身,不过在 x 月 x 日因为觉得太累而放弃了。"

CLI 命令

安装后,可以通过以下方式调用:

# 直接运行(启动 MCP Server stdio 模式)
talk-to-me

# 或使用 npx(无需安装)
npx -y @lakphy/talk-to-me

注意: MCP Server 通过 stdio 协议通信,通常由 MCP 客户端(如 Claude Desktop、Cursor)自动调用,不需要手动运行。

本地开发

# 克隆项目
git clone <repo-url>
cd talk-to-me

# 安装依赖
pnpm install

# 开发模式(监听文件变化)
pnpm dev

# 构建
pnpm build

# 类型检查
pnpm typecheck

# 测试 CLI
./scripts/test-cli.sh

本地测试

如需本地测试,在 MCP 配置中使用绝对路径:

{
  "mcpServers": {
    "talk-to-me": {
      "command": "node",
      "args": ["/Users/your-username/projects/talk-to-me/dist/index.mjs"]
    }
  }
}

License

MIT