Token导航 LogoToken导航

Agent Skill · 研究检索

curiosity-gap(好奇心差距)

curiosity-gap 用于查找、检索和筛选相关信息,适合在 Codex、Claude、Cursor、Gemini CLI 中需要根据关键词、任务场景或来源线索快速定位候选结果时使用。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。

curiosity-gap 是研究检索类 Agent Skill,可用宿主为 Codex、Claude、Cursor、Gemini CLI,通过GitHub 仓库获取,权限为只读;许可证信息还不明确,第三方扫描已通过,按安装量在研究检索类的 73,675 个 Skill 里排第 53,392。

总安装 535周安装 23GitHub Stars 235下载 188核验 2026-05-01

安装说明

本站只整理中文说明,不托管安装包,也不代用户安装。
GitHub
许可证
unknown

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:curiosity-gap(好奇心差距)
来源仓库:https://github.com/flpbalada/my-opencode-config
仓库路径:skills/curiosity-gap
安装命令:
npx skills add https://github.com/flpbalada/my-opencode-config --skill curiosity-gap
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/flpbalada/my-opencode-config --skill curiosity-gap

这个 Skill 是做什么的

研究检索类 Skill 主要解决「找资料、核对事实、把来源整理清楚」的问题,通常会让助手先检索再总结。

用于查找、检索和筛选相关信息,适合在 Codex、Claude、Cursor、Gemini CLI 中快速定位候选结果。可结合关键词、任务场景或来源线索进行信息筛选。使用时需要明确目标用户群体和应用场景,区分不同领域。安装前建议确认权限范围和维护状态,避免触发联网或文件读写操作。

注意该技能侧重心理学理论应用,不适合技术实现层面的具体工作。

来源仓库内路径:skills/curiosity-gap

上手步骤

4 步
  1. 1

    确认自己的宿主工具

    公开资料标注支持 Codex、Claude、Cursor、Gemini CLI;先确认自己用的是其中之一,再按对应工具的 Skills 目录放置。

  2. 2

    拿到安装命令

    可以直接复制的命令是 npx skills add https://github.com/flpbalada/my-opencode-config --skill curiosity…;命令里的占位符换成自己的路径或密钥再执行。

  3. 3

    安装后重启宿主

    命令执行完要完全退出宿主工具再打开,让新装的 Skill 被重新扫描到,否则列表里可能看不到。

  4. 4

    按最小权限试用

    它标注为只读,先用只读任务试一次,确认行为符合预期再扩大使用范围。

常见问题

6 条

curiosity-gap 怎么安装?

在终端执行 npx skills add https://github.com/flpbalada/my-opencode-config --skill curiosity-gap。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

curiosity-gap 支持哪些 AI 工具?

公开资料标注支持 Codex、Claude、Cursor、Gemini CLI。其他工具只要能读取本地 Skill 目录,一般也能手动放进去,但要自己确认触发方式是否一致。

curiosity-gap 需要什么权限?

权限级别是只读。该 Skill 主要提供规则、说明或参考内容,本身偏只读;真正读写文件、联网或执行命令仍取决于宿主 Agent 的任务。

curiosity-gap 是免费的吗?

公开资料没有标注明确许可证,收费和商用限制都无法确认。正式使用前请到来源仓库核对授权说明。

curiosity-gap 用的人多吗?

公开数据是累计安装 535 次,近一周安装 23 次,下载 188 次,GitHub 235 星。这些数字来自第三方目录统计,只能当作热度参考,不代表质量或安全性。

curiosity-gap 安全吗?

第三方扫描结果为通过,但这只覆盖已知的静态检查项,不能替代自己读一遍源码和安装命令。

值得选的地方和要先想清楚的地方

值得选的地方

  • 同时适配 4 种工具,换客户端不用重新找替代品。
  • 第三方扫描已通过,常见的高风险行为没有命中。
  • 标注为只读权限,试用阶段不容易误改本地文件。

先想清楚的地方

  • 许可证没有明确标注,能不能商用、能不能再分发都不确定。
  • 第三方目录整理,版本更新和行为细节以来源仓库为准。

适不适合你

更适合

  • 已经在用 Codex,想直接扩展它能力的人。
  • 需要处理研究检索类任务、又不想每次都手写提示词的人。
  • 只想先试试效果、不希望它改动本地文件的人。

不太适合

  • 需要明确授权才能用到公司项目里的场景。
  • 需要厂商级技术支持和 SLA 的生产系统。

研究检索类 Skill 对比

6 个同类
Skill总安装GitHub 星权限许可证扫描
find-skills

寻找技能 · Codex、Claude

31,199,81316,486操作浏览器MIT有提醒
self-improving-agent

自我改进 Agent · OpenClaw

16,464,0023,396接触敏感数据MIT-0通过
soultrace

灵魂追踪 · Codex、Claude

7,588,94412需要联网未明确有提醒
self-improving

自我完善 · OpenClaw

6,829,2841,046需要联网MIT-0通过
microsoft-foundry

微软代工厂 · Codex、Claude

6,614,144764调用外部服务未明确有提醒
curiosity-gap

好奇心差距 · 本页

535235只读未明确通过

按安装量取同分类前几名;数字来自第三方目录统计,只作热度参考,不代表质量结论。

SKILL.md 原文

Curiosity Gap - Psychology of Information Seeking

The curiosity gap is a psychological phenomenon where people experience an unpleasant feeling of uncertainty when aware of missing information, creating intense motivation to fill that knowledge gap. Based on George Loewenstein's Information Gap Theory, curiosity functions like a drive state—similar to hunger—compelling people to seek resolution.

When to Use This Skill

  • Writing headlines and subject lines
  • Designing notification strategies
  • Creating onboarding and feature discovery flows
  • Planning content marketing
  • Building engagement loops
  • Designing product reveals and launches

Psychological Foundation

Information Gap Theory (Loewenstein, 1994):

Knowledge Gap Detected
        ↓
Cognitive Tension (uncomfortable uncertainty)
        ↓
Drive-State Activation (like hunger)
        ↓
Attention Focus (on missing information)
        ↓
Action to Resolve (click, read, explore)
        ↓
Reward (caudate nucleus activation)

Neuroscience

fMRI studies show curiosity activates:

  • Caudate nucleus: Reward anticipation
  • Inferior frontal gyrus: Information seeking
  • Same regions activated by monetary rewards

Curiosity is literally rewarding to resolve.

Curiosity Gap Anatomy

Effective Curiosity Gap Structure:

┌─────────────────────────────────────────────────────┐
│                                                      │
│   REVEAL enough to:        CONCEAL enough to:       │
│   ├── Establish relevance  ├── Create uncertainty   │
│   ├── Build interest       ├── Trigger motivation   │
│   └── Signal value         └── Drive action         │
│                                                      │
│   Balance Point: User knows WHAT but not HOW/WHY    │
│                                                      │
└─────────────────────────────────────────────────────┘

Gap Types

TypeTriggerExample
Knowledge gap"I don't know this""The strategy Netflix uses to..."
Outcome gap"What happened?""She tried it, and then..."
Opinion gap"What do others think?""Experts are divided on..."
Resolution gap"How does this end?""The surprising conclusion..."

Application Framework

Step 1: Define the Payoff

Before creating a gap, know what you're promising:

Payoff Clarity:

What will the user gain?
├── Information (learn something new)
├── Solution (solve a problem)
├── Entertainment (experience something)
└── Status (know what others don't)

The gap must lead to real value, not disappointment.

Step 2: Calibrate the Gap

Gap Size Spectrum:

Too Small                          Too Large
"Click to see"   ←───────────→   "Mind-blowing secret"
     ↓                                    ↓
No motivation                      Skepticism/distrust

Sweet Spot:
├── Specific enough to be believable
├── Intriguing enough to motivate
└── Relevant to user's interests

Step 3: Choose the Reveal Strategy

StrategyWhen to UseExample
ImmediateHigh-value contentClick → full article
ProgressiveOnboarding/educationUnlock next lesson
SocialCommunity featuresSee what friends shared
TimedAnticipation buildingReveal tomorrow at noon

Output Template

After designing curiosity gaps, document as:

## Curiosity Gap Design

**Context:** [Where this will be used]

**Date:** [Date]

### Gap Analysis

| Element        | Current    | Improved      | Gap Type                 |
| -------------- | ---------- | ------------- | ------------------------ |
| [Headline/CTA] | [Original] | [New version] | [Knowledge/Outcome/etc.] |

### Payoff Validation

| Gap     | Promised Value      | Delivered Value | Match? |
| ------- | ------------------- | --------------- | ------ |
| [Gap 1] | [What user expects] | [What they get] | ✅/❌  |

### Ethical Checklist

- [ ] Gap leads to genuine value
- [ ] Promise matches delivery
- [ ] Not exploiting vulnerable states
- [ ] User can easily access information
- [ ] Not creating unnecessary anxiety

### Success Metrics

| Metric                   | Target | Measurement    |
| ------------------------ | ------ | -------------- |
| Click-through rate       | X%     | [How measured] |
| Time to resolution       | X sec  | [How measured] |
| Satisfaction post-reveal | X/10   | [How measured] |
| Return engagement        | X%     | [How measured] |

Real-World Examples

Netflix: Series Previews

Strategy:

Teaser trailers (30 sec):
├── Show: Dramatic moments, character intrigue
├── Hide: Plot resolution, key reveals
└── Gap: "What happens next?"

Episode thumbnails:
├── Show: Emotional peak moment
├── Hide: Context and resolution
└── Gap: "How did they get there?"

Result: 15% increase in viewing engagement

Duolingo: Streak Psychology

Strategy:

Streak display:
├── Show: "47-day streak! 🔥"
├── Gap: "What happens if I break it?"
└── Drive: Loss aversion + curiosity

Achievement previews:
├── Show: Locked badge silhouette
├── Hide: Requirements and rewards
└── Gap: "What do I need to unlock this?"

Result: 65% DAU complete lessons even when unmotivated

Spotify: Discover Weekly

Strategy:

Weekly playlist:
├── Show: "Your Discover Weekly is ready"
├── Hide: Actual song selections
├── Gap: "What did the algorithm find?"
└── Timing: Every Monday (anticipation)

Wrapped campaigns:
├── Show: "Your 2024 Wrapped is ready"
├── Hide: Personal statistics
├── Gap: "What are my listening habits?"
└── Social: "What did others get?"

Result: 40M hours/week listening time

Ethical Guidelines

✅ Do: Create Value-Driven Gaps

Ethical Applications:

Progressive disclosure:
├── Reveal information in digestible layers
├── User controls the pace
└── Each reveal provides value

Feature discovery:
├── Highlight unused capabilities
├── Show benefit before asking exploration
└── Easy to dismiss if not interested

Content previews:
├── Accurate representation of full content
├── Clear path to resolution
└── Value matches or exceeds expectation

❌ Don't: Manipulate or Mislead

Dark Patterns to Avoid:

Clickbait:
├── "You won't believe..." → Mundane content
├── "This one trick..." → Obvious information
└── Promise ≠ Delivery

False urgency:
├── "Last chance!" → Perpetual availability
├── "Only 2 left!" → Unlimited inventory
└── Manufactured scarcity

Emotional manipulation:
├── Fear-based gaps exploiting anxiety
├── FOMO that creates unhealthy behavior
└── Guilt-inducing incompleteness

The Ethics Test

Before deploying a curiosity gap, ask:

1. Does the payoff match the promise?
   YES → Continue
   NO  → Revise or abandon

2. Would I feel satisfied as the user?
   YES → Continue
   NO  → Revise payoff

3. Is this building trust or eroding it?
   BUILDING → Continue
   ERODING  → Stop

4. Can users easily get resolution?
   YES → Continue
   NO  → Remove barriers

Headline Patterns

Knowledge Gap Headlines

Templates:

"How [Group] [Achieve Result]"
→ "How top performers stay focused for 8+ hours"

"The [Adjective] way to [Outcome]"
→ "The counterintuitive way to learn faster"

"What [Authority] knows about [Topic]"
→ "What neuroscientists know about decision fatigue"

"Why [Common Belief] is wrong"
→ "Why multitasking doesn't save time"

Outcome Gap Headlines

Templates:

"[Person/Company] tried [Action]. Here's what happened."
→ "We removed all meetings for a month. Here's what happened."

"[X] [Action], and then [Incomplete]..."
→ "She asked for a raise, and then..."

"The unexpected result of [Action]"
→ "The unexpected result of saying no to everything"

Measurement

Key Metrics

MetricWhat It ShowsTarget
Click-through rateGap effectivenessBaseline +X%
Time to clickGap urgencyLower is better
Completion ratePayoff satisfaction>80%
Return rateTrust maintainedIncreasing
Satisfaction scorePromise delivery>4/5

Warning Signs

Gap is too weak if:
├── CTR below baseline
├── High bounce immediately after click
└── No engagement increase

Gap is too strong (clickbait) if:
├── High CTR but low satisfaction
├── Complaints about misleading content
├── Decreasing return engagement
└── Trust metrics declining

Integration with Other Methods

MethodCombined Use
Cognitive BiasesCuriosity gap leverages multiple biases
Loss Aversion"Don't miss" framing strengthens gaps
Social Proof"See what others discovered"
Progressive DisclosureStructured gap resolution
Hooked ModelCuriosity as trigger/reward

Quick Reference

CURIOSITY GAP CHECKLIST

Creating the Gap:
□ Payoff clearly defined
□ Gap size calibrated (not too small/large)
□ Relevant to target audience
□ Specific and believable

Ensuring Ethics:
□ Promise matches delivery
□ User feels satisfied after resolution
□ No exploitation of vulnerable states
□ Easy path to resolution

Measuring Success:
□ CTR tracking in place
□ Satisfaction measured post-reveal
□ Return engagement monitored
□ Trust metrics watched

Resources

平台分布

Codex

32.62%
按下载量换算61

Claude

29.25%
按下载量换算55

Cursor

19.74%
按下载量换算37

Gemini CLI

10.35%
按下载量换算19

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

通过

权限和风险

只读

该 Skill 主要提供规则、说明或参考内容,本身偏只读;真正读写文件、联网或执行命令仍取决于宿主 Agent 的任务。

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。目前只有一个公开来源,建议同时核对源仓库。

继续浏览同类 Skills