Token导航 LogoToken导航

Agent Skill · 研究检索

fogg-behavior-model(雾行为模型)

fogg-behavior-model 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。

fogg-behavior-model 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 48,626。

总安装 682周安装 29GitHub Stars 239下载 239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:fogg-behavior-model(雾行为模型)
来源仓库:https://github.com/flpbalada/my-opencode-config
仓库路径:skills/fogg-behavior-model
安装命令:
npx skills add https://github.com/flpbalada/my-opencode-config --skill fogg-behavior-model
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 fogg-behavior-model

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。

fogg-behavior-model 用于查找、检索和筛选相关信息。适合在 Codex、Claude、Cursor、Gemini CLI 中根据任务场景快速定位信息。可结合来源仓库和原始 README 核验具体用法。安装方式:github,通过 npx skills add 命令添加。

注意:需确认权限范围和维护状态,避免触发不必要的联网或文件操作。

来源仓库内路径:skills/fogg-behavior-model

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 fogg-beha…;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

fogg-behavior-model 怎么安装?

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

fogg-behavior-model 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。

fogg-behavior-model 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。

fogg-behavior-model 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。

fogg-behavior-model 用的人多吗?

公开数据是累计安装 682 次,近一周安装 29 次,下载 239 次,GitHub 239 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。

fogg-behavior-model 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
fogg-behavior-model

雾行为模型 · 本页

682239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

Fogg Behavior Model - B = MAP

The Fogg Behavior Model explains that three elements must converge at the same moment for a behavior to occur: Motivation, Ability, and a Prompt. When a behavior does not occur, at least one of these elements is missing.

When to Use This Skill

  • Designing onboarding and activation flows
  • Improving conversion rates
  • Building habit-forming products
  • Increasing feature adoption
  • Understanding why users drop off
  • Planning behavior change interventions

The B = MAP Formula

┌─────────────────────────────────────────────────────────────────┐
│                    BEHAVIOR = MAP                                │
│                                                                  │
│    Behavior happens when Motivation, Ability, and Prompt        │
│    come together at the SAME MOMENT.                            │
│                                                                  │
│    When behavior doesn't happen → at least one is missing.      │
└─────────────────────────────────────────────────────────────────┘

    High  │                    ·····
    M     │              ·····     Behavior
    o     │        ·····           Happens
    t     │   ·····                Here
    i     │····─────────────────────────────
    v     │   Action Line
    a     │
    t     │        Behavior
    i     │        Fails
    o     │        Here
    n     │
    Low   └─────────────────────────────────────
              Hard ←── Ability ──→ Easy

    Prompts only work above the Action Line.

The Three Elements

1. Motivation

What drives the user to act?

Motivation Sources:

Core Motivators (Fogg):
├── Pleasure / Pain
├── Hope / Fear
└── Social Acceptance / Rejection

Additional Drivers:
├── Intrinsic interest
├── Personal goals
├── External rewards
└── Social pressure
MotivatorLowHigh
Pleasure/Pain"I should exercise""I want to feel great"
Hope/Fear"Might be useful""Don't want to miss out"
Social"No one cares""Everyone's doing it"

2. Ability

How easy is it to do?

Ability Factors (Fogg):

Simplicity Chain (weakest link determines ability):
├── Time: How long does it take?
├── Money: How much does it cost?
├── Physical effort: How hard physically?
├── Mental effort: How much thinking?
├── Social deviance: How weird is it?
└── Non-routine: How different from habits?

Ability = Inverse of the HARDEST factor
FactorLow AbilityHigh Ability
Time30-minute signup2-click signup
Money$99/monthFree trial
PhysicalVisit storeClick button
MentalComplex formSmart defaults
SocialPublic commitmentPrivate action
RoutineNew behaviorFits existing habit

3. Prompt

What triggers action at the right moment?

Prompt Types:

Spark (High Ability, Low Motivation):
├── Inspires and motivates
├── Appeals to emotions
└── Example: "Your friends are waiting"

Facilitator (High Motivation, Low Ability):
├── Makes action easier
├── Reduces friction
└── Example: "One-click purchase"

Signal (High Motivation, High Ability):
├── Simple reminder
├── Just needs timing
└── Example: "Time to check in"

Behavior Diagnosis Framework

Step 1: Define Target Behavior

Be specific about what you want users to do:

Behavior Definition:

❌ Vague: "Use the app more"
✅ Specific: "Complete a 5-minute workout daily"

Components:
├── Who: [Target user segment]
├── What: [Specific action]
├── When: [Timing/context]
└── How often: [Frequency]

Step 2: Diagnose Missing Element

Diagnosis Tree:

Is the user doing the behavior?
│
├── NO → Diagnose which element is missing:
│   │
│   ├── Do they WANT to do it?
│   │   ├── NO → Motivation problem
│   │   └── YES → Continue
│   │
│   ├── CAN they easily do it?
│   │   ├── NO → Ability problem
│   │   └── YES → Continue
│   │
│   └── Are they PROMPTED at the right moment?
│       ├── NO → Prompt problem
│       └── YES → Re-examine all three
│
└── YES → Behavior successful

Step 3: Design Intervention

ProblemSolution Approach
Low MotivationIncrease desire (spark prompt)
Low AbilityReduce friction (facilitator prompt)
No PromptAdd well-timed trigger (signal prompt)
Multiple issuesStart with Ability (easiest to change)

Output Template

After completing analysis, document as:

## Behavior Design Analysis

**Target Behavior:** [Specific behavior]

**User Segment:** [Who]

**Date:** [Date]

### Current State

| Element    | Score (1-5) | Evidence              |
| ---------- | ----------- | --------------------- |
| Motivation | [Score]     | [What indicates this] |
| Ability    | [Score]     | [What indicates this] |
| Prompt     | [Score]     | [What indicates this] |

### Ability Breakdown

| Factor   | Current State | Bottleneck? |
| -------- | ------------- | ----------- |
| Time     | [Assessment]  | Yes/No      |
| Money    | [Assessment]  | Yes/No      |
| Physical | [Assessment]  | Yes/No      |
| Mental   | [Assessment]  | Yes/No      |
| Social   | [Assessment]  | Yes/No      |
| Routine  | [Assessment]  | Yes/No      |

### Diagnosis

**Primary Issue:** [Motivation/Ability/Prompt]

**Root Cause:** [Specific reason]

### Intervention Design

| Priority | Change            | Element | Expected Impact      |
| -------- | ----------------- | ------- | -------------------- |
| 1        | [Specific change] | [M/A/P] | [Measurable outcome] |
| 2        | [Specific change] | [M/A/P] | [Measurable outcome] |

### Success Metrics

| Metric          | Current | Target | Timeline |
| --------------- | ------- | ------ | -------- |
| [Behavior rate] | X%      | Y%     | [Time]   |

Real-World Examples

Example 1: Daily Exercise Habit

Target Behavior: Do a 20-minute workout daily

Motivation:
├── Want to get fit ✓
├── Feel better ✓
└── Score: 4/5 (High)

Ability:
├── Time: 20 min → Could be less
├── Physical: Moderate effort
├── Mental: Need to decide what to do
├── Routine: Not part of current habits
└── Score: 2/5 (Low - bottleneck)

Prompt:
├── No consistent trigger
└── Score: 2/5 (Low)

Interventions:
├── Ability: Reduce to 5-minute starter routine
├── Ability: Pre-select workout (no decisions)
├── Prompt: Phone alarm + clothes laid out
└── Routine: Anchor to morning coffee

Example 2: Feature Adoption (SaaS)

Target Behavior: Use new collaboration feature

Motivation:
├── Users don't see value yet
└── Score: 2/5 (Low - problem)

Ability:
├── Feature is buried in menu
├── Requires 4 clicks to access
└── Score: 2/5 (Low - problem)

Prompt:
├── One email announcement sent
└── Score: 1/5 (Very low)

Interventions:
├── Motivation: Show social proof ("Teams save 2hrs/week")
├── Ability: Add one-click access from dashboard
├── Ability: Pre-configure with defaults
├── Prompt: In-app tooltip at relevant moment
└── Prompt: Contextual suggestion during related tasks

Example 3: Newsletter Signup

Target Behavior: Subscribe to weekly newsletter

Motivation:
├── Valuable content promised
├── Social proof: "10,000 subscribers"
└── Score: 3/5 (Medium)

Ability:
├── Email only (simple)
├── One field
└── Score: 5/5 (High)

Prompt:
├── Popup after 30 seconds
├── User often not ready yet
└── Score: 2/5 (Wrong timing)

Intervention:
├── Prompt: Move to end of valuable article
├── Prompt: "Want more like this?"
└── Context: After user received value

Design Principles

Start with Ability

Why Ability First:

Motivation:
├── Hard to change
├── Often outside your control
└── Fluctuates over time

Ability:
├── Directly designable
├── Permanent once improved
└── Helps when motivation dips

"Make it so easy they can't say no."

Right Prompt, Right Moment

Prompt Timing:

Too Early:
├── User not ready
├── Creates annoyance
└── Wasted impression

Too Late:
├── Moment passed
├── Motivation cooled
└── Friction accumulated

Just Right:
├── High motivation moment
├── Ability is present
└── Action is natural next step

Tiny Habits Approach

BJ Fogg's Tiny Habits:

1. Make it TINY
   └── Smallest possible version of behavior

2. Find the right ANCHOR
   └── Existing habit to attach to

3. Celebrate IMMEDIATELY
   └── Positive emotion reinforces

Formula: "After I [ANCHOR], I will [TINY BEHAVIOR]"

Example: "After I pour my coffee, I will do 2 pushups"

Behavior Types

TypeMotivationAbilityFocus
GreenHighHighJust add prompt
BlueHighLowIncrease ability
PurpleLowHighIncrease motivation
GrayLowLowMajor redesign needed

Integration with Other Methods

MethodCombined Use
Hooked ModelFogg explains the trigger/action phase
Cognitive LoadAbility = inverse of cognitive load
Loss AversionPowerful motivation lever
Curiosity GapMotivation through information gaps
Five WhysWhy isn't behavior happening?

Quick Reference

B = MAP CHECKLIST

Motivation Boosters:
□ Clear value proposition
□ Social proof present
□ Loss framing considered
□ Personalized relevance
□ Emotional connection

Ability Enhancers:
□ Minimum steps possible
□ Smart defaults set
□ No unnecessary fields
□ Mobile-friendly
□ Fits existing routines

Prompt Optimization:
□ Right type for situation
□ Appears at right moment
□ Clear call to action
□ Not interruptive
□ Contextually relevant

Resources

平台分布

Codex

36.23%
按下载量换算87

Claude

32.77%
按下载量换算78

Cursor

17.7%
按下载量换算42

Gemini CLI

9.2%
按下载量换算22

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