Token导航 LogoToken导航TokenDH.com
研究检索需要联网github未标认证来源可访问许可证需确认审计通过

thinking-effectuation思维实现

Agent Skill

thinking-effectuation 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。

总安装

423

周安装

18

GitHub Stars

46

下载量

148
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:thinking-effectuation(思维实现)
来源仓库:https://github.com/tjboudreaux/cc-thinking-skills
仓库路径:skills/thinking-effectuation
安装命令:
npx skills add https://github.com/tjboudreaux/cc-thinking-skills --skill thinking-effectuation
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/tjboudreaux/cc-thinking-skills --skill thinking-effectuation

简介

思维实现用于查找、检索和筛选相关信息,强调从可控资源出发构建可能性。

  • 适用于创业规划、创新项目设计等不确定性较高的探索型任务。
  • 以现有能力为基础,推演可实现的路径而非预测未来趋势。
  • 安装命令:npx skills add https://github.com/tjboudreaux/cc-thinking-skills --skill thinking-effectuation
  • 需留意其逻辑推导是否隐含假设,建议人工复核关键结论。

SKILL.md

Effectuation

Overview

Effectuation, developed by Saras Sarasvathy from studying expert entrepreneurs, inverts traditional causal reasoning. Instead of starting with a goal and finding resources, effectuators start with available means and discover goals through action. This approach is more robust in highly uncertain environments where prediction is unreliable.

Core Principle: When you can't predict the future, don't try. Instead, focus on what you can control and let the future emerge from your actions.

When to Use

  • Startup and new product development
  • Innovation in uncertain domains
  • When prediction is unreliable
  • Entering new markets
  • When resources are constrained
  • Side projects and experiments
  • Career pivots

Decision flow:

Facing uncertainty about outcomes?
  → Can you predict the future reliably? → no → USE EFFECTUATION
  → Do you have fixed goals but uncertain resources? → yes → Traditional planning
  → Do you have available means but uncertain goals? → yes → USE EFFECTUATION

Causal vs Effectual Reasoning

Causal (Traditional)

1. Set a goal
2. Plan to achieve the goal
3. Gather necessary resources
4. Execute the plan
5. Measure against the goal

Example:
Goal: Build a $10M company
Plan: Build product X for market Y
Resources: Raise $2M, hire 10 people
Execute: 18-month development plan
Measure: Revenue against projections

Effectual (Entrepreneurial)

1. Start with your means
2. Take action with acceptable loss
3. Build partnerships
4. Leverage contingencies
5. Goals emerge from action

Example:
Means: I know databases, have some savings, know other engineers
Action: Build a small tool, see who's interested
Partners: Three early users want to co-create
Contingency: One user has a different problem that's bigger
Emerging goal: Pivot to new problem, user becomes co-founder

The Five Principles of Effectuation

1. Bird in Hand (Start with Means)

Begin with what you have:

## My Means Inventory

Who I am:
- Skills: [List skills]
- Preferences: [What I enjoy]
- Values: [What matters to me]

What I know:
- Domain expertise: [Areas of knowledge]
- Unique insights: [What I see that others don't]
- Technical skills: [What I can build]

Who I know:
- Potential partners: [People who might join]
- Potential customers: [People who might buy]
- Resources: [People who can help]

What can I do with THESE means?
(Don't start with "what's needed"—start with "what I have")

2. Affordable Loss (Acceptable Downside)

Focus on what you can afford to lose, not expected return:

## Affordable Loss Analysis

I can afford to lose:
- Time: 6 months of evenings/weekends
- Money: $5K of savings
- Opportunity: Delay of other projects
- Reputation: Minor if it fails quietly

I cannot afford to lose:
- Day job income
- Family time beyond X hours
- More than $5K

Action: Design the experiment to fit within affordable loss
        Don't calculate expected return—calculate maximum loss
        Can I live with maximum loss? If yes, proceed.

3. Crazy Quilt (Partnerships)

Co-create with anyone who commits:

## Partnership Building

Instead of: Finding resources for my predetermined plan
Do: Let partners shape the venture

Approach:
1. Share what I'm working on
2. Anyone who commits becomes a partner
3. They bring their means and constraints
4. The venture adapts to include them

Example:
- I start with: Database tool idea
- Partner 1 commits: Brings sales experience, shifts to B2B
- Partner 2 commits: Brings design, shifts to user-facing product
- Customer 1 commits: Brings specific use case, shapes roadmap

The venture becomes what committed partners make it.

4. Lemonade (Leverage Contingencies)

Treat surprises as opportunities:

## Contingency Response

Causal mindset: "That wasn't in the plan—it's a problem"
Effectual mindset: "That's unexpected—how can we use it?"

Examples:
- Competitor launches similar product
  Causal: "Our plan is threatened"
  Effectual: "They validated market, potential acquirer, potential partner"

- Key hire falls through
  Causal: "We're behind plan"
  Effectual: "Maybe we don't need that role; what can we do with who we have?"

- Customer wants something different
  Causal: "That's not our product"
  Effectual: "Maybe that IS our product"

5. Pilot in the Plane (Control, Don't Predict)

Focus on what you can control:

## Control vs Predict

Can't control/predict:
- Market movements
- Competitor actions
- Customer adoption rate
- Economic conditions
- Technology shifts

Can control:
- Who I work with
- What I build this week
- How I respond to feedback
- What partnerships I form
- How much I risk

Strategy: Maximize actions on controllable factors
          Don't waste energy predicting uncontrollable factors
          Create the future through action, not prediction

Effectuation Process

Step 1: Inventory Your Means

## Means Inventory

### Who I Am
- Background: [Experience, skills, identity]
- Passions: [What energizes me]
- Values: [What I won't compromise]

### What I Know
- Professional knowledge: [Domains]
- Technical skills: [Abilities]
- Market insights: [What I understand that others might not]

### Who I Know
- Close network: [People who would take my call]
- Extended network: [People who know people]
- Communities: [Groups I'm part of]

Step 2: Define Affordable Loss

## Affordable Loss

Time I can invest: [Hours/week, months]
Money I can lose: [$X]
Reputation at stake: [What's the downside?]
Opportunity cost: [What else am I not doing?]

Maximum I'm willing to lose: [Clear boundary]

Step 3: Take Action

## Next Action

Given my means and affordable loss:
What's the smallest action that could create information?

Options:
1. [Action] - Teaches me [what]
2. [Action] - Teaches me [what]
3. [Action] - Teaches me [what]

Selected action: [Action with best learning/risk ratio]

Step 4: Seek Commitments

## Partnership Outreach

Who might be interested in co-creating?

| Person | Their Means | Ask | Commitment |
|--------|-------------|-----|------------|
| [Name] | [Skills/resources] | [What to ask] | [What they commit] |

Each commitment expands my means and shapes the direction.

Step 5: Adapt and Iterate

## Iteration Log

| Action | Outcome | Surprise | How to leverage |
|--------|---------|----------|-----------------|
| | | | |

After each action:
- What did I learn?
- What commitments emerged?
- What contingencies can I leverage?
- What's the next action?

Effectuation Template

# Effectuation Analysis: [Venture/Project]

## Means Inventory
Who I am: [Identity, skills, values]
What I know: [Knowledge, expertise]
Who I know: [Network, relationships]

## Affordable Loss
Can afford: [Time, money, reputation]
Cannot afford: [Boundaries]
Maximum loss I'm willing to accept: [Clear number]

## Current Commitments
| Partner | Their Means | Commitment |
|---------|-------------|------------|
| | | |

## Contingencies Available
| Surprise | Potential Leverage |
|----------|-------------------|
| | |

## Next Action
Action: [Specific, small action]
Affordable loss: [What's at stake]
Learning goal: [What I'll know after]

## Emerging Direction
Where seems to be going: [Current trajectory]
How it differs from start: [Evolution]

When to Use Causal vs Effectual

ContextApproachWhy
Existing market, proven modelCausalPrediction is reliable
New market, unproven modelEffectualPrediction is unreliable
Abundant resourcesCausalCan afford to plan extensively
Constrained resourcesEffectualMust work with what you have
Clear goalCausalPlan toward the goal
Exploring opportunitiesEffectualGoals emerge from action
Risk can be calculatedCausalExpected return is meaningful
Risk is uncertainEffectualAffordable loss is meaningful

Verification Checklist

  • Inventoried available means
  • Defined affordable loss (not expected return)
  • Sought commitments, not just resources
  • Treating surprises as opportunities
  • Focusing on what I can control
  • Taking action to learn, not to execute plan
  • Letting goals emerge from action

Key Questions

  • "What can I do with what I have?"
  • "What am I willing to lose?"
  • "Who might want to co-create this?"
  • "How can I leverage this surprise?"
  • "What can I control right now?"
  • "What's the smallest action that teaches me something?"

Sarasvathy's Wisdom

"Effectual reasoning does not begin with a specific goal. Instead, it begins with a given set of means and allows goals to emerge contingently over time from the varied imaginations and diverse aspirations of the founders and the people they interact with."

"In the face of an uncertain future, entrepreneurs use effectual logic to fabricate—make—the future, rather than try to find or predict it."

You can't predict the future. But you can create it through action. Start with your means, take affordable risks, build with partners, leverage surprises, and control what you can. The goal will find you.

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

能力 4

展示第三方安全扫描或审计结果

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

32.91%
按下载量换算49

Claude

30.48%
按下载量换算45

Cursor

19.42%
按下载量换算29

Gemini CLI

8.72%
按下载量换算13

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

通过

权限和风险

需要联网

该 Skill 可能需要联网访问来源站点、仓库或外部 API;具体网络访问范围需要结合源码和 README 复核。

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