Token导航 LogoToken导航TokenDH.com
待分类权限需确认github未标认证来源可访问许可证需确认审计未展示

minimum-lovable-product-(mlp)最小可爱产品(mlp)

Agent Skill

minimum-lovable-product-(mlp) 用于处理 GitHub 仓库、Issue、Pull Request 和代码协作信息,适合在 Codex、Claude、Cursor、Gemini CLI 中需要围绕仓库状态、代码变更或协作事项进行整理时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

总安装

6,940

周安装

211

GitHub Stars

2

下载量

1,861
CodexClaudeCursorGemini CLI

安装说明

本站只整理中文说明和来源信息,不托管安装包,也不代用户安装。

GitHub

来源数

2

许可证

unknown

最后核验

2026-05-01

来源状态

来源可访问

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:minimum-lovable-product-(mlp)(最小可爱产品(mlp))
来源仓库:https://github.com/coowoolf/insighthunt-skills
仓库路径:skills/minimum-lovable-product-(mlp)
安装命令:
npx skills add https://github.com/coowoolf/insighthunt-skills --skill 'Minimum Lovable Product (MLP)'
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。该命令会通过 npx skills 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

skills.shnpx skills
npx skills add https://github.com/coowoolf/insighthunt-skills --skill 'Minimum Lovable Product (MLP)'

简介

用于处理 GitHub 仓库、Issue、Pull Request 和代码协作信息。

  • 适合围绕仓库状态、代码变更或协作事项进行整理。
  • 可结合来源仓库和原始 README 继续核验具体用法。
  • 安装方式:github,通过 npx skills add 命令从 coowoolf/insighthunt-skills 仓库添加。
  • 适用于 Codex、Claude、Cursor、Gemini CLI 等宿主环境。

SKILL.md

The Minimum Lovable Product (MLP)

"Viability is left back in 2010s. Now it's minimal lovable product." — Elena Verna

What It Is

A development standard that prioritizes emotional connection, brand feel, and "magic" over simple functional viability. In an era where AI lowers the cost of building software, the differentiator becomes the joy of use.

When To Use

  • Entering a crowded market with low barriers to entry
  • Competitors offer similar utility
  • When "MVP market tests" yield indifference, not feedback
  • Building consumer or prosumer products

MVP vs. MLP

         MVP (2010s)                    MLP (Now)

    ✓ Does it work?                ✓ Does it delight?
    ✓ Is it viable?                ✓ Is it lovable?
    ✓ Aha moment                   ✓ Wow moment
    ✓ Understand value             ✓ Feel the magic

    Result: "It's okay"            Result: "I have to share this!"

Core Principles

1. Brand is Product Interaction

Do not separate brand marketing from product design. Every UI interaction must convey the brand's personality.

2. Prioritize "Wow" Over "Aha"

The goal is an immediate feeling of "I can't believe this is possible" (Wow), not just understanding value (Aha).

3. Fix "Unlovable" Bugs Immediately

If a feature works but feels clunky or lifeless, treat it as a P0 bug. Stop the line to fix the "vibe."

4. Designers as Early Hires

Hire high-agency designers earlier than usual to ensure the emotional layer is built alongside the functional layer.

How To Apply

STEP 1: Define Your "Lovable" Standard
└── What makes users share this?
└── What creates word-of-mouth?

STEP 2: Audit for Vibes
└── Does every interaction feel magical?
└── Is the personality consistent?

STEP 3: Create "Lovable Veto"
└── Anyone can flag feature as "not lovable"
└── Team drops sprint to fix it immediately

STEP 4: Measure Emotional Response
└── NPS is not enough
└── Look for unprompted shares and social posts

Common Mistakes

❌ Shipping a dry, utilitarian MVP that yields indifference

❌ Separating "brand" from "product" (two different teams)

❌ Treating emotional polish as a post-PMF luxury

Real-World Example

Lovable's internal culture where anyone can flag a feature as "not lovable," causing the team to drop sprints to fix the interaction immediately.


*Source: Elena Verna, Head of Growth at Lovable, Lenny's Podcast*

适合场景

01

用户想查找某类 Agent Skill 时

02

需要根据任务场景推荐可安装能力包时

03

需要对比不同来源的安装命令和来源信息时

能力概览

能力 1

按任务关键词查找相关 Skills

能力 2

展示可复制的安装命令

能力 3

保留来源站点、仓库和原始说明,方便继续核验

安装后应在对应宿主中按原始 README 的触发条件使用;具体调用方式请以来源页面和 README 为准。

平台分布

Codex

34.76%
按下载量换算647

Claude

33.69%
按下载量换算627

Cursor

18.06%
按下载量换算336

Gemini CLI

9.63%
按下载量换算179

安全审计

暂无安全审计结果可展示。

权限和风险

权限需确认

当前来源未能明确判断权限范围,默认进入异常复核队列。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。当前只有一个来源,正式发布前建议补源仓库或其他目录站核验。

来源信息

继续浏览同类 Skills