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

@wayne.evan/v5-memory

v1.1.0

Published

V5 Meta Memory System - Universal LLM Memory Framework

Readme

V5 元记忆系统 (V5 Memory Meta-System)

通用大模型全局记忆框架 | V5 Meta Structure Powered

核心定位

一句话:让 AI 记住用户的一切,实现跨平台、跨场景的个性化记忆能力。

技术核心:V5 势垒方程 P = 1/(1+e^(-2γ(Input-B))) 驱动记忆的提取、召回、注入全生命周期。

方程本质:Sigmoid 函数变体,将「输入匹配度」映射为 0-1 之间的「记忆激活概率」,实现记忆召回/写入的量化决策。


三大版本

| 版本 | 定位 | 用户群体 | 技术栈 | 核心差异 | |------|------|----------|--------|----------| | 完整版 | 全功能、全平台适配 | 高级用户、研究者 | Node.js CLI + Web | 支持向量检索、云端同步、自定义元规则 | | 网页版 | 浏览器插件形态 | 写手、视觉设计师 | Chrome Extension (Manifest V3) | 轻量本地存储、仅关键词检索、主流 LLM 适配 | | IDE版 | 开发工具形态 | 程序员、架构师 | VS Code 插件 / Electron | 适配 IDE 内 AI 助手(Cursor/Windsurf)、代码上下文记忆 |


功能矩阵

| 功能 | 完整版 | 网页版 | IDE版 | 实现说明 | |------|--------|--------|-------|----------| | 多平台适配 | ✅ | ✅ | ✅ | Meta-Adapt Layer | | 自动记忆提取 | ✅ | ✅ | ✅ | V5 势垒方程阈值过滤 | | 智能召回注入 | ✅ | ✅ | ✅ | V5 势垒方程排序 | | V5 元规则引擎 | ✅ | ✅ | ✅ | 核心引擎驱动 | | 记忆中心 UI | ✅ | ✅ | ✅ | 筛选/编辑记忆 | | 向量语义检索 | ✅ (可选) | ❌ | ❌ | LocalEmbedding | | 云端同步 | ✅ (可选) | ❌ | ❌ | 加密同步 | | 自定义角色 | ✅ | ✅ | ✅ | namespace 隔离 |


V5 元结构

┌─────────────────────────────────────────────────────────────┐
│                      V5 Meta Engine                        │
├─────────────┬─────────────┬─────────────┬──────────────────┤
│  Meta-Adapt │  Meta-Opera  │  Meta-Data  │   Meta-Security │
│   Layer    │   Layer     │   Layer     │     Layer       │
│  平台适配  │  记忆操作   │  元数据管理 │    安全管控     │
├─────────────┴─────────────┴─────────────┴──────────────────┤
│                     Storage Layer                          │
│           (IndexedDB / localStorage / FileSystem)          │
└─────────────────────────────────────────────────────────────┘

核心组件

  1. Meta-Adapt Layer - 平台适配器:DeepSeek/ChatGPT/Claude/Gemini/Cursor/Windsurf/Cline
  2. Meta-Opera Layer - 提取/召回/注入器 + 规则引擎
  3. Meta-Data Layer - 版本链、关联关系、生命周期
  4. Meta-Security Layer - 敏感信息、脱敏、审计、隔离

关键技术实现

V5 势垒方程

export function v5BarrierEquation(input, gamma = 0.85, barrier = 0.5) {
  const normalizedInput = Math.max(0, Math.min(1, input));
  const exponent = -2 * gamma * (normalizedInput - barrier);
  return parseFloat((1 / (1 + Math.exp(exponent))).toFixed(4));
}

// input=0.7 → P=0.8176 (高概率召回)
// input=0.4 → P=0.3820 (低概率召回)

写入得分

export function calculateWriteScore(confidence, importance, platformWeight = 1, freshness = 1) {
  return (confidence * 0.4) + (importance * 0.3) + (platformWeight * 0.2) + (freshness * 0.1);
}

记忆类型阈值

| 类型 | 阈值 | 说明 | |------|------|------| | pinned | 1.0 | 置顶(最高) | | persona | 0.8 | 用户画像 | | core | 0.6 | 核心记忆 | | episodic | 0.4 | 情境细节 |


快速开始

完整版

cd src/complete
npm install
node cli.js --init
node cli.js --input "我喜欢Python" --platform deepseek
node cli.js --retrieve "推荐框架"

网页版

cd src/web/frontend
npm install && npm run build
# 加载 dist 到 Chrome

IDE版

cd src/ide/vscode
npm install
# F5 调试

项目结构

v5-memory/
├── README.md
├── SPEC.md
├── src/
│   ├── core/
│   │   ├── engine/
│   │   │   ├── meta_engine.js   # 核心引擎
│   │   │   ├── scorer.js        # V5 势垒方程
│   │   │   ├── extractor.js     # 记忆提取
│   │   │   ├── retriever.js    # 记忆召回
│   │   │   └── injector.js     # 记忆注入
│   │   ├── adapt/
│   │   │   ├── base_adapter.js
│   │   │   └── platform_adapters.js
│   │   ├── security/
│   │   │   └── sanitizer.js    # 敏感信息+审计
│   │   └── storage/
│   │       └── memory_store.js
│   ├── complete/cli.js
│   ├── web/frontend/
│   └── ide/vscode/

合规声明

  • 技术框架,不涉及玄学/占卜
  • 本地存储优先,隐私优先
  • 敏感信息自动脱敏
  • 一键清空功能

许可证

MIT - 望易 V5 元结构体系