Xpersona
agentCLAWHUBUnverified

Feynman Think Method

运用费曼学习法和物理思维，深入理解复杂概念，识别盲点，构建第一性原理认知，助力学习与问题诊断。 Skill: Feynman Think Method Owner: gechengling Summary: 运用费曼学习法和物理思维，深入理解复杂概念，识别盲点，构建第一性原理认知，助力学习与问题诊断。 Tags: feynman-think-method:2.1.2, latest:2.1.2 Version history: v2.1.2 | 2026-09-21T05:13:06.442Z | user 2.1.2: 四模型各补实战举例与对照表；新增费曼理解度自评表（7维度5列）；动态更新至2026-09-21并新增分析列；补充适用边界与数据最小化声明 v2.1.1 | 2026-06-28T13:21:18.371Z | auto - 新增“最新动态”和“投资思维最新动态”板块，展示AI技术应用与金融投资宏观框架的实时行业更新。 - 补充2026年最新达利欧、AI治理等事件，深化AI时代知识与投资风险提示。 - 丰富

OpenClaw
View on ClawHubHomepage

Rank

62

Safety

84

Downloads

1.1k

Updated

Oct 11, 2026

Version

2.1.2

Source

CLAWHUB

About

What it does, and when to use it.

Capability contract not published. No trust telemetry is available yet. 1.1K downloads reported by the source. Last updated 10/11/2026.

Avoid when

  • Contract metadata is missing or unavailable for deterministic execution.

Risk flags: missing_or_unavailable_contract, trust_data_unavailable, schema_references_missing

Public facts

Every fact links back to the source it came from.

Vendor
Clawhubvendor · observed Oct 11, 2026
Protocol compatibility
OpenClawcompatibility · observed Oct 11, 2026
Adoption signal
1.1K downloadsadoption · observed Oct 11, 2026
Latest release
2.1.2release · observed Sep 21, 2026
Handshake status
UNKNOWNsecurity

Install and run

Setup complexity: low.

clawhub skill install s17ewqc4f2s6gpcbm88hy7fgvn85kg1g:feynman-think-method
  1. Setup complexity is LOW. This package is likely designed for quick installation with minimal external side-effects.
  2. Final validation: Expose the agent to a mock request payload inside a sandbox and trace the network egress before allowing access to real customer data.

Contract: missing

curl -s "https://www.xpersona.co/api/v1/agents/clawhub-gechengling-feynman-think-method/snapshot"

Documentation

CLAWHUB

27,894 characters of source documentation, loaded on request.

Extracted files

3 files captured from the source.

SKILL.md

# SKILL.md



## Identity

- **Skill Name**: 费曼学习法与物理学思维 (Richard Feynman Think Method)

- **Slug**: finance-feynman-learning

- **Version**: 2.1.2
- **Updated**: 2026-09-21
- **适用边界（Scope Boundary）**：本技能面向**概念理解与方法论训练**，输出的是思考路径与检验问题，不是事实性结论。涉及具体投资标的、医疗诊断、法律责任的判断，本技能只负责帮你把问题问清楚，结论仍须由具备资质的专业人员在核实原始资料后作出。
- **数据最小化（Data Minimization）**：使用本技能时请只提供理解该概念所必需的最小信息；涉及客户身份、健康、财务、未公开经营数据的内容，请先脱敏或改写为抽象表述再输入。本技能不需要也不索取任何凭据、密钥、账号信息。

- **Language**: 中文为主，英文关键术语保留

- **Author**: 葛成 (@gechengling)

- **Description**: 以费曼学习法和物理学思维为核心，帮助深入理解复杂概念、识别真正知识盲点、建立第一性原理式认知。适用于学习研究、问题诊断、知识管理等场景。2026更新：新增费曼方法在AI时代的应用——"向AI学习如何提问"与"让AI扮演费曼检验你的理解"的对话范式。关键词：费曼学习法，物理学思维，第一性原理，知识理解，认知升级，简单原则，教学相长.



### 最新动态 [2026-09-21更新]

| 动态类型 | 内容摘要 | 发布时间 | 影响范围 | 费曼式追问（对本技能的启示） |
|---------|---------|---------|---------|---------------------------|
| AI价值审视 | AI技术“实际渗透率<20%”与“资本市场狂热定价”的矛盾，费曼式“回到第一性原理”审视AI真实价值 | 2026-06 | AI研究/投资 | 追问第1层：这家公司的收入里，有多少来自AI真实付费，多少来自预算试点？ |
| AI治理落地 | 银行业保险业人工智能安全开发应用相关指导意见持续落地，金融机构AI应用从“能用”走向“可解释、可审计” | 2026-06-18 | 金融机构AI应用与内控 | 追问第2层：模型的每一个输出，能否追到输入数据与责任人？追不到就不算理解 |
| 大模型能力迭代 | 主流大模型在长链推理、工具调用与多模态理解上持续迭代，“向AI提问”的质量成为新的能力分水岭 | 2026-Q3 | 学习方法与知识管理 | 追问第3层：你问出的问题，是AI替你想的，还是你自己想问的？ |
| 认知科学补充 | “解释深度错觉”（Illusion of Explanatory Depth）研究在教育与培训场景被反复验证：自评理解度与实际可复述度显著背离 | 2026-Q3 | 学习方法与培训设计 | 追问第4层：把解释写出来之前，你并不知道自己不会——这正是费曼法存在的理由 |



> **数据截止**: 2026-09-21 | 来源：AI行业研究报告、国家金融监督管理总局公开信息、公开学术文献
> **声明**: 以上动态供参考，具体以官方最新发布为准



## Core Thinking Models



### 模型一：费曼学习法四步

```

Step 1: 选择一个概念，写下你知道的一切

Step 2: 假装教给一个12岁孩子 → 卡住的地方=真正盲点

Step 3: 回头填补缺口，重新研究直到流畅

Step 4: 简化+类比 → 找不到类比=理解不够深

```



**举例 A：用四步法拆解“久期（Duration）”**



| 步骤 | 你实际会写下的内容 | 卡住的地方（真盲点） |
|------|-------------------|---------------------|
| Step 1 | 久期是债券现金流的加权平均回收时间，用来衡量利率风险 | 写不出“加权”的权重到底是什么 |
| Step 2 | 对12岁孩子说：“借钱给别人，他早还还是晚还，你的感觉不一样” | 说不清“为什么价格变动≈久期×利率变动” |
| Step 3 | 回去看定义：权重是各期现金流现值占总现值的比例 | 原来关键是“现值”，不是“金额” |
| Step 4 | 类比：久期像一根扁担的支点位置，支点越远，一头动一点另一头晃得越厉害 | — |



**举例 B：用四步法拆解“偿付能力充足率”**



| 步骤 | 你实际会写下的内容 | 卡住的地方（真盲点） |
|------|-------------------|---------------------|
| Step 1 | 实际资本除以最低资本，监管要求不低于100% | 说不出“实际资本”和会计净资产差在哪 |
| Step 2 | 对12岁孩子说：“保险公司要留够一笔钱，保证出事也赔得起” | 讲不清为什么“最低资本”是风险敞口的函数 |
| Step 3 | 回到规则：最低资本由保险风险、市场风险、信用风险分别计量后加总 | 分母是“风险”，不是“规模” |
| Step 4 | 类比：不是看你家有多少存款，而是看万一同时来三场大病，够不够付 | — |



### 模型二：第一性理解

- 第一类知识：知道名称 ≠ 真正理解
- 能用类比解释？能预测陌生场景表现？能教给完全不懂的人？
- 费曼：“如果你不能向一个12岁孩子解释清楚，你就没有真正理解它”



**类比检验的三道关（举例）**



| 概念 | 类比尝试 | 类比能预测陌生场景吗？ | 判定 |
|------|---------|----------------------|------|
| 注意力机制（Attention） | 像开会时你只认真听跟自己议题相关的人发言 | 能：可以推出“上下文越长，单条信息的权重越被稀释” | 通过，真理解 |
| 注意力机制（Attention） | 像聚光灯照在舞台上 | 不能：推不出多头注意力为什么要并行多束光 | 未通过，只记住了名字 |
| 利差损 | 像你按5%的利息借钱，却只能投出3%的收益 | 能：可以推出“负债成本刚性+资产收益率下行=亏损扩大” | 通过，真理解 |
| 利差损 | 像做生意亏本 | 不能：推不出为什么久期错配会放大亏损 | 未通过，只是贴标签 |



### 模型三：追问本质

- “为什么？” 连续追问5层 → 剥离表象直达物理现实
- 费曼：“我宁愿有无法回答的问题，也不愿有无法质疑的答案”
- 证伪比证实更重要



**举例：把“这家寿险公司利差损

_meta.json

{
  "ownerId": "kn74e704j3ygjcygnpf02rdvd185js13",
  "slug": "feynman-think-method",
  "version": "2.1.2",
  "publishedAt": 1789967586442
}

skill-card.md

## Description:

Guides agents to help users understand complex concepts through Feynman-style explanation, analogy, first-principles questioning, and understanding checks.

This skill is ready for commercial/non-commercial use.

## Publisher:

[gechengling](https://clawhub.ai/user/gechengling)

### License/Terms of Use:

MIT-0

## Use Case:

Developers and end users can use this skill to turn vague or partial understanding into teachable explanations, identify knowledge gaps, and structure follow-up research. It is most appropriate for learning, research preparation, problem diagnosis, and knowledge-management workflows rather than professional advice or real-time factual lookup.

### Deployment Geography for Use:

Global

## Known Risks and Mitigations:

Risk: Users may mistake conceptual reasoning support for verified financial, legal, medical, or current-market advice.

Mitigation: Frame outputs as learning aids and require qualified professional review plus source verification before relying on them for regulated or high-impact decisions.

Risk: Users may provide personal, client, health, financial, or confidential business data while asking for examples or diagnosis.

Mitigation: Ask for minimized, desensitized, or abstracted inputs and avoid requesting credentials, account details, or sensitive identifiers.

Risk: The artifact includes time-sensitive examples and external claims that may become stale or unverified.

Mitigation: Treat dates, figures, laws, market claims, and citations as prompts for source checking rather than authoritative facts.

## Reference(s):

- [ClawHub skill page](https://clawhub.ai/gechengling/skills/feynman-think-method)
- [Publisher profile](https://clawhub.ai/user/gechengling)

## Skill Output:

**Output Type(s):** [text, markdown, guidance]

**Output Format:** [Markdown or structured text with explanations, questions, examples, comparison tables, and self-check rubrics]

**Output Parameters:** [1D]

**Other Properties Related to Output:** [Produces reasoning aids and learning prompts; does not produce verified professional advice, real-time facts, credentials, code execution, or persistent state.]

## Skill Version(s):

2.1.2 (source: server evidence release.version and artifact identity)

## Ethical Considerations:

Users should evaluate whether this skill is appropriate for their environment, review any generated or modified files before relying on them, and apply their organization's safety, security, and compliance requirements before deployment.

Related agents

Same protocol, similar job.

Github ReposUpdated 2d agoRank 70

AionUi

Free, local, open-source 24/7 Cowork app and OpenClaw for Gemini CLI, Claude Code, Codex, OpenCode, Qwen Code, Goose CLI, Auggie, and more | 🌟 Star if you like it!

MCPOPENCLAW
Github ReposUpdated 6mo agoRank 70

activepieces

AI Agents & MCPs & AI Workflow Automation • (~400 MCP servers for AI agents) • AI Automation / AI Agent with MCPs • AI Workflows & AI Agents • MCPs for AI Agents

OPENCLAW
Github ReposUpdated 6mo agoRank 70

cherry-studio

AI productivity studio with smart chat, autonomous agents, and 300+ assistants.

MCPOPENCLAW
Github ReposUpdated 7mo agoRank 70

CopilotKit

The Frontend for Agents & Generative UI. React + Angular

OPENCLAW

Machine-readable data

The same record, as JSON, for agents and crawlers.

{
  "facts": [
    {
      "factKey": "vendor",
      "category": "vendor",
      "label": "Vendor",
      "value": "Clawhub",
      "href": "https://clawhub.ai/gechengling/skills/feynman-think-method",
      "sourceUrl": "https://clawhub.ai/gechengling/skills/feynman-think-method",
      "sourceType": "profile",
      "confidence": "medium",
      "observedAt": "2026-10-11T08:03:09.411Z",
      "isPublic": true
    },
    {
      "factKey": "protocols",
      "category": "compatibility",
      "label": "Protocol compatibility",
      "value": "OpenClaw",
      "href": "https://www.xpersona.co/api/v1/agents/clawhub-gechengling-feynman-think-method/contract",
      "sourceUrl": "https://www.xpersona.co/api/v1/agents/clawhub-gechengling-feynman-think-method/contract",
      "sourceType": "contract",
      "confidence": "medium",
      "observedAt": "2026-10-11T08:03:09.411Z",
      "isPublic": true
    },
    {
      "factKey": "traction",
      "category": "adoption",
      "label": "Adoption signal",
      "value": "1.1K downloads",
      "href": "https://clawhub.ai/gechengling/feynman-think-method",
      "sourceUrl": "https://clawhub.ai/gechengling/feynman-think-method",
      "sourceType": "profile",
      "confidence": "medium",
      "observedAt": "2026-10-11T08:03:09.411Z",
      "isPublic": true
    },
    {
      "factKey": "latest_release",
      "category": "release",
      "label": "Latest release",
      "value": "2.1.2",
      "href": "https://clawhub.ai/gechengling/feynman-think-method",
      "sourceUrl": "https://clawhub.ai/gechengling/feynman-think-method",
      "sourceType": "release",
      "confidence": "medium",
      "observedAt": "2026-09-21T05:13:06.442Z",
      "isPublic": true
    },
    {
      "factKey": "handshake_status",
      "category": "security",
      "label": "Handshake status",
      "value": "UNKNOWN",
      "href": "https://www.xpersona.co/api/v1/agents/clawhub-gechengling-feynman-think-method/trust",
      "sourceUrl": "https://www.xpersona.co/api/v1/agents/clawhub-gechengling-feynman-think-method/trust",
      "sourceType": "trust",
      "confidence": "medium",
      "observedAt": null,
      "isPublic": true
    }
  ],
  "events": [
    {
      "eventType": "release",
      "title": "Release 2.1.2",
      "description": "2.1.2: 四模型各补实战举例与对照表；新增费曼理解度自评表（7维度5列）；动态更新至2026-09-21并新增分析列；补充适用边界与数据最小化声明",
      "href": "https://clawhub.ai/gechengling/feynman-think-method",
      "sourceUrl": "https://clawhub.ai/gechengling/feynman-think-method",
      "sourceType": "release",
      "confidence": "medium",
      "observedAt": "2026-09-21T05:13:06.442Z",
      "isPublic": true
    }
  ]
}

Record generated Oct 11, 2026.

Sponsored

Ads related to Feynman Think Method and adjacent AI workflows.