Agent Skills › MaxKmet/idea-validation-agents › desire-evaluator

desire-evaluator

GitHub

评估应用创意对五大核心人类欲望（生存、地位、归属、控制、好奇）的契合度，预测用户留存与传播潜力。输入概念描述，输出各维度得分及整体强度标签。

skills/desire-evaluator/SKILL.md MaxKmet/idea-validation-agents

Trigger Scenarios

分析新应用创意的市场吸引力评估产品是否具备高留存和病毒式传播特征判断用户需求驱动的强弱

Install

npx skills add MaxKmet/idea-validation-agents --skill desire-evaluator -g -y

More Options

Use without installing

npx skills use MaxKmet/idea-validation-agents@desire-evaluator

指定 Agent (Claude Code)

npx skills add MaxKmet/idea-validation-agents --skill desire-evaluator -a claude-code -g -y

安装 repo 全部 skill

npx skills add MaxKmet/idea-validation-agents --all -g -y

预览 repo 内 skill

npx skills add MaxKmet/idea-validation-agents --list

SKILL.md

Frontmatter

{
    "name": "desire-evaluator",
    "description": "Scores the strength of core human desire motivations (survival, status, belonging, control, curiosity) for a given app idea to predict user pull and retention potential."
}

Skill: desire-evaluator

Purpose

Apps that tap into primal human desires outperform apps that only solve functional problems. This skill scores how strongly an app idea connects to core motivational drivers, which predicts organic virality, retention, and pricing power.

Input

Idea slug
App concept description
Optional: memory/ideas/<slug>/user_extraction.json (identity driver)
Use this file as reference of what drives core human desires memory/extra-context/core-human-desires.md

Desire Dimensions

Dimension	Description	Example App
Survival	Health, safety, financial security	Calorie tracker, budgeting app
Status	Looking good, achieving, winning	Fitness leaderboard, portfolio tracker
Belonging	Community, connection, not being alone	Group savings, running clubs
Control	Mastery, autonomy, reducing chaos	Task manager, habit tracker
Curiosity	Learning, discovery, novelty	Language app, quiz game

Process

Score each dimension 1–5 based on how directly the app taps into it.
Identify the primary and secondary desire drivers.
Compute overall desire strength (weighted average).
Flag if no dimension scores ≥ 3 (weak desire signal → high churn risk).

Output

Write to memory/ideas/<slug>/desire_scores.json:

{
  "scores": {
    "survival": 0,
    "status": 0,
    "belonging": 0,
    "control": 0,
    "curiosity": 0
  },
  "primary_driver": "",
  "secondary_driver": "",
  "desire_strength": 0,
  "desire_strength_label": "strong | moderate | weak",
  "virality_potential": "high | medium | low",
  "notes": ""
}