Token导航 LogoToken导航TokenDH.com
研究检索权限需确认clawhub未标认证来源可访问clear审计通过

analogical-reasoning类比推理

Agent Skill

analogical-reasoning 用于查找、检索和筛选相关信息,适合在 OpenClaw 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

10,992

周安装

458

GitHub Stars

公开资料未说明

下载量

3,664
OpenClaw

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

MIT-0

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:analogical-reasoning(类比推理)
来源仓库:https://github.com/wanikua/analogical-reasoning
安装命令:
openclaw skills install analogical-reasoning
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 OpenClaw 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。

ClawHubOpenClaw
openclaw skills install analogical-reasoning

简介

analogical-reasoning 应用类比推理方法,将已有知识迁移至新领域以辅助创造性问题解决。

  • 适用于跨领域创新、复杂系统理解或结构相似性分析等场景。
  • 通过 clawhub 安装后,可在 OpenClaw 中调用其逻辑映射能力支持决策过程。
  • 建议结合具体案例验证推理结果的适用性与准确性后再实施行动。
  • 适用宿主包括 OpenClaw,接入前应确认版本、权限和运行环境要求。

SKILL.md

name
analogical-reasoning
description
Apply analogical reasoning to transfer knowledge from familiar domains to unfamiliar ones. Use when the user needs creative problem-solving by finding structural parallels, wants to understand something new through comparison, or needs to evaluate whether an analogy holds or breaks down.

Analogical Reasoning

Analogical reasoning transfers knowledge from a familiar domain (the "source") to an unfamiliar one (the "target") by identifying structural similarities. It's how humans naturally make sense of the new — by connecting it to the known. Used brilliantly by scientists (Rutherford: atom is like a solar system), entrepreneurs (Uber for X), and legal scholars (case law precedent). But analogies can also mislead when surface similarities mask deep structural differences. The key is knowing when the mapping holds and when it breaks.


Analyze the current topic or problem under discussion using analogical reasoning. Find illuminating parallels, map them carefully, and extract transferable insights — while being honest about where the analogy breaks down. Apply this framework to whatever the user is currently working on or asking about.


Step 1: Understand the Target Domain

*First, deeply understand the problem you're trying to solve.*

  • What are the key elements of this problem? (Actors, relationships, dynamics, constraints, goals)
  • What makes this problem hard? Where is the core difficulty?
  • What is unknown or uncertain about this domain?
  • What structure underlies the problem? (Causal relationships, feedback loops, trade-offs)
  • Temporarily set aside domain-specific details — focus on the abstract structure.

Step 2: Generate Source Analogies

*Find domains that share structural features with your target.*

Search broadly across domains. For each, briefly state the analogy:

Near Analogies (same general field)

  • What similar problem in a related domain has been solved before?
  • What does the nearest competitor or adjacent industry do?

Far Analogies (completely different fields)

  • Nature/Biology: What organism, ecosystem, or evolutionary process mirrors this?
  • History: What historical event or era parallels this situation?
  • Engineering/Physics: What physical system behaves similarly?
  • Games/Sports: What game or sporting strategy has this structure?
  • Medicine: What medical condition or treatment protocol is analogous?
  • Military: What military strategy or campaign matches?
  • Art/Music: What creative process or composition mirrors this?
  • Economics: What market or economic phenomenon has the same dynamics?

Generate at least 5 source analogies, with at least 2 from distant domains. Far analogies are often more creative and insightful than near ones.

Step 3: Deep Mapping — Structure the Best Analogies

For the top 3 most promising analogies, perform a detailed structural mapping:

Analogy: [Source Domain] → [Target Domain]

Source ElementTarget ElementMapping Strength
[Actor/component in source][Corresponding actor in target]Strong/Moderate/Weak
[Relationship in source][Corresponding relationship]Strong/Moderate/Weak
[Dynamic/process in source][Corresponding dynamic]Strong/Moderate/Weak
[Constraint in source][Corresponding constraint]Strong/Moderate/Weak
[Outcome in source][Predicted outcome in target]Strong/Moderate/Weak

Key questions for each mapping:

  • Is the correspondence structural (deep) or merely surface (superficial)?
  • Is the causal mechanism the same, or just the appearance?
  • Does the mapping scale appropriately?

Step 4: Extract Transferable Insights

For each strong analogy:

  • What solutions or strategies worked in the source domain?
  • What principles underlie those solutions (not the specific details — the abstract principles)?
  • How would those principles translate to the target domain?
  • What predictions does the analogy make about the target? (These are testable!)
  • What pitfalls were discovered in the source domain that the target should avoid?
  • What timeline or trajectory did the source domain follow? Does the target follow a similar path?

Step 5: Identify Where the Analogy Breaks Down

*This is the most important step. All analogies are wrong; some are useful.*

For each analogy:

  • Where do the structural correspondences fail?
  • What key features of the target domain have no counterpart in the source?
  • What features of the source domain are irrelevant or misleading in the target?
  • Where does the analogy predict something false about the target?
  • What is the disanalogy — the most important difference?
  • How might relying on this analogy lead you astray?

Rate the analogy's overall reliability:

  • High fidelity: Core structure maps well, breakdowns are in peripheral details
  • Medium fidelity: Structure partially maps, some important differences
  • Low fidelity: Surface similarity only, deep structure differs significantly

Step 6: Triangulate Across Analogies

  • Where do multiple analogies converge on the same insight? (High confidence)
  • Where do they diverge? (Indicates complexity or important nuance)
  • What insight appears in the far analogies that's invisible in the near ones?
  • What composite analogy (combining elements from multiple sources) best captures the target?

Step 7: Generate Novel Solutions

Based on the analogical analysis:

  • What specific solutions or approaches from the source domains could be adapted?
  • What novel combination of source-domain strategies creates something new?
  • What would a practitioner from the source domain suggest if they saw this problem?
  • What experiment or test would validate whether the analogical transfer actually works?

Synthesis

  • State the most illuminating analogy and the key insight it provides.
  • Acknowledge the limits of the analogy explicitly.
  • Recommend concrete actions inspired by the analogy, adjusted for where it breaks down.

George Pólya said: "Analogy pervades all our thinking." The art is not in finding analogies — the human mind does that instinctively. The art is in testing them rigorously: mapping the structure, checking the correspondence, and being honest about where the parallel fails.

适合场景

01

OpenClaw 用户查找和安装 Skill 时

02

用户想查找某类 Agent Skill 时

03

需要根据任务场景推荐可安装能力包时

04

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

补充不同宿主或平台的使用分布数据

能力 5

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

OpenClaw

70.41%
按下载量换算2,580

安全审计

VirusTotal

通过

ClawScan

通过

Static analysis

通过

权限和风险

权限需确认

当前来源未能明确判断权限范围,默认进入异常复核队列。

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