Token导航 LogoToken导航

Agent Skill · 研究检索

socratic-teaching-scaffolds(苏格拉底式教学支架)

socratic-teaching-scaffolds 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。

socratic-teaching-scaffolds 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,权限为需要联网;许可证为 MIT,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 28,750。

总安装 3,166周安装 136GitHub Stars 85下载 1,110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
MIT

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:socratic-teaching-scaffolds(苏格拉底式教学支架)
来源仓库:https://github.com/lyndonkl/claude
仓库路径:skills/socratic-teaching-scaffolds
安装命令:
npx skills add https://github.com/lyndonkl/claude --skill socratic-teaching-scaffolds
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。不同来源提供的安装方式可能略有差异;本站展示可直接复制的安装命令,安装前请核对来源页面。

skills.shnpx skills
npx skills add https://github.com/lyndonkl/claude --skill socratic-teaching-scaffolds

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。

socratic-teaching-scaffolds 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。面向教育工作者和学习者设计,提供模块化教学支持工具以辅助概念讲解与技能训练。

可根据学习目标动态调整问题难度与反馈节奏,实现个性化学习路径推荐。安装命令为 npx skills add https://github.com/lyndonkl/claude --skill socratic-teaching-scaffolds,需评估其对本地资源访问的需求。

建议检查项目活跃度与社区评价,防止因废弃维护导致功能异常或兼容中断。

来源仓库内路径:skills/socratic-teaching-scaffolds

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/lyndonkl/claude --skill socratic-teaching-scaf…;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

    它需要需要联网,先在测试目录或测试账号里跑,确认没问题再放到正式项目。

常见问题

6 条

socratic-teaching-scaffolds 怎么安装?

在终端执行 npx skills add https://github.com/lyndonkl/claude --skill socratic-teaching-scaffolds。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

socratic-teaching-scaffolds 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。

socratic-teaching-scaffolds 需要什么权限?

权限级别是需要联网。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。

socratic-teaching-scaffolds 是免费的吗?

来源标注的许可证是 MIT。本站不代收任何费用,也不托管安装包;能否商用、要不要署名,以许可证原文为准。

socratic-teaching-scaffolds 用的人多吗?

公开数据是累计安装 3,166 次,近一周安装 136 次,下载 1,110 次,GitHub 85 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。

socratic-teaching-scaffolds 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。
  • 3 个目录同时收录,说明这个 Skill 流传比较广。
  • 许可证明确为 MIT,商用和分发边界好判断。
  • 第三方扫描已通过,常见的高风险行为没有命中。

先想清楚的地方

  • 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。
  • 需要许可证清楚、并且有第三方扫描结果的团队项目。

不太适合

  • 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
socratic-teaching-scaffolds

苏格拉底式教学支架 · 本页

3,16685需要联网MIT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

Socratic Teaching Scaffolds

Table of Contents

  1. Workflow
  2. Socratic Question Types
  3. Scaffolding Levels
  4. Common Patterns
  5. Guardrails
  6. Quick Reference

Core components:

  1. Question Ladders: Sequences from simple to complex that build understanding incrementally
  2. Misconception Detectors: Questions that reveal faulty mental models through contradiction
  3. Feynman Explanations: Build-up from simple analogies to technical precision
  4. Worked Examples with Fading: Full solutions → partial solutions → independent practice
  5. Cognitive Apprenticeship: Model thinking process explicitly, then transfer to learner

Quick example (Teaching Recursion):

Question Ladder:

  1. "Can you break this problem into a smaller version of itself?" (problem decomposition)
  2. "What would happen if we had only one item?" (base case discovery)
  3. "If we could solve the small version, how would we use it for the big version?" (recursive case)
  4. "What prevents this from running forever?" (termination reasoning)

Misconception Detector:

  • "Will this recursion ever stop? Trace it with 3 items." (reveals infinite recursion misunderstanding)

Feynman Progression:

  • Level 1: "Like Russian nesting dolls--each contains a smaller version"
  • Level 2: "Function calls itself with simpler input until base case"
  • Level 3: "Recursive definition: f(n) = g(f(n-1), n) with f(0) = base"

Workflow

Copy this checklist and track your progress:

Socratic Teaching Progress:
- [ ] Step 1: Diagnose learner's current understanding
- [ ] Step 2: Design question ladder and scaffolding plan
- [ ] Step 3: Guide discovery through questioning
- [ ] Step 4: Fade scaffolding as competence grows
- [ ] Step 5: Validate understanding and transfer

Step 1: Diagnose learner's current understanding

Ask probing questions to identify current knowledge level, misconceptions, and learning goals. See Socratic Question Types for diagnostic question categories.

Step 2: Design question ladder and scaffolding plan

Build progression from learner's current state to target understanding. For straightforward teaching → Use resources/template.md. For complex topics with multiple misconceptions → Study resources/methodology.md.

Step 3: Guide discovery through questioning

Ask questions in sequence, provide scaffolding (hints, worked examples, analogies) as needed. See Scaffolding Levels for support gradations. Adjust based on learner responses.

Step 4: Fade scaffolding as competence grows

Progressively remove hints, provide less complete examples, ask more open-ended questions. Monitor for struggle (optimal challenge) vs frustration (too hard). See resources/methodology.md for fading strategies.

Step 5: Validate understanding and transfer

Test with novel problems, ask for explanations in learner's words, check for misconception elimination. Self-check using resources/evaluators/rubric_socratic_teaching_scaffolds.json. Minimum standard: Average score ≥ 3.5.

Socratic Question Types

1. Clarifying Questions (Understand current thinking)

  • "What do you mean by [term]?"
  • "Can you give me an example?"
  • "How does this relate to [known concept]?"

2. Probing Assumptions (Surface hidden beliefs)

  • "What are we assuming here?"
  • "Why would that be true?"
  • "Is that always the case?"

3. Probing Reasons/Evidence (Justify claims)

  • "Why do you think that?"
  • "What evidence supports that?"
  • "How would we test that?"

4. Exploring Implications (Think through consequences)

  • "What would happen if [change]?"
  • "What follows from that?"
  • "What are the edge cases?"

5. Questioning the Question (Meta-cognition)

  • "Why is this question important?"
  • "What are we really trying to understand?"
  • "How would we know if we understood?"

6. Revealing Contradictions (Bust misconceptions)

  • "Earlier you said [X], but now [Y]. How do these fit?"
  • "If that's true, why does [counterexample] happen?"
  • "What would this predict for [test case]?"

Scaffolding Levels

Provide support that matches current need, then fade:

Level 5: Full Modeling (I do, you watch)

  • Complete worked example with thinking aloud
  • Explicit strategy articulation
  • All steps shown with rationale

Level 4: Guided Practice (I do, you help)

  • Partial worked example
  • Ask learner to complete steps
  • Provide hints before errors

Level 3: Coached Practice (You do, I help)

  • Learner attempts independently
  • Intervene with questions when stuck
  • Guide without giving answers

Level 2: Independent with Feedback (You do, I watch)

  • Learner solves alone
  • Review and discuss afterwards
  • Identify gaps for next iteration

Level 1: Transfer (You teach someone else)

  • Learner explains to others
  • Learner creates examples
  • Learner identifies misconceptions in others

Fading strategy: Start at level matching current competence (not Level 5 by default). Move down one level when learner demonstrates success. Move up one level if learner struggles repeatedly.

Common Patterns

Pattern 1: Concept Introduction (Concrete → Abstract)

  • Start: Real-world analogy or example
  • Middle: Formalize with terminology
  • End: Abstract definition with edge cases
  • Example: Teaching pointers (address on envelope → memory location → pointer arithmetic)

Pattern 2: Misconception Correction (Prediction → Surprise → Explanation)

  • Ask learner to predict outcome
  • Show actual result (contradicts misconception)
  • Guide discovery of correct mental model
  • Example: "Will this float? [test with 0.1 + 0.2 in programming] Why not exactly 0.3?"

Pattern 3: Problem-Solving Strategy (Model → Practice → Reflect)

  • Model strategy on simple problem (think aloud)
  • Learner applies to similar problem (with coaching)
  • Reflect on when strategy applies/fails
  • Example: Teaching debugging (print statements → breakpoints → hypothesis testing)

Pattern 4: Depth Ladder (ELI5 → Undergraduate → Expert)

  • Build multiple explanations at different depths
  • Let learner choose starting point
  • Provide "go deeper" option at each level
  • Example: Teaching neural networks (pattern matching → weighted sums → backpropagation → optimization theory)

Pattern 5: Discovery Learning (Puzzle → Hints → Insight)

  • Present puzzling phenomenon or problem
  • Provide graduated hints if stuck
  • Guide to "aha" moment of discovery
  • Example: Teaching recursion (Towers of Hanoi → break into subproblems → recursive solution)

Guardrails

Zone of proximal development:

  • Too easy = boredom, too hard = frustration
  • Optimal: Can't do alone, but can with guidance
  • Adjust scaffolding level based on struggle signals

Don't fish for specific answers:

  • Socratic questioning isn't a guessing game
  • If learner's reasoning is sound but reaches different conclusion, explore their path
  • Multiple valid approaches often exist

Avoid pseudo-teaching:

  • Don't just ask questions without purpose
  • Each question should advance understanding or reveal misconception
  • If question doesn't help, provide direct explanation

Misconception resistance:

  • Deep misconceptions resist single corrections
  • Need multiple exposures to contradictions
  • May require building correct model from scratch before dismantling wrong one

Expertise blind spots:

  • Experts forget what was hard as beginners
  • Make implicit knowledge explicit
  • Slow down automated processes to show thinking

Individual differences:

  • Some learners prefer exploration, others prefer structure
  • Adjust scaffolding style to learner preferences
  • Monitor for frustration vs productive struggle

Quick Reference

Resources:

5-Step Process: Diagnose → Design Ladder → Guide Discovery → Fade Scaffolding → Validate Transfer

Question Types: Clarifying, Probing Assumptions, Probing Evidence, Exploring Implications, Meta-cognition, Revealing Contradictions

Scaffolding Levels: Full Modeling → Guided Practice → Coached Practice → Independent Feedback → Transfer (fade progressively)

Patterns: Concrete→Abstract, Prediction→Surprise→Explanation, Model→Practice→Reflect, ELI5→Expert, Puzzle→Hints→Insight

Guardrails: Zone of proximal development, purposeful questions, avoid pseudo-teaching, resist misconceptions, make implicit explicit

平台分布

Claude Code

28.2%
按下载量换算313

Gemini CLI

24.88%
按下载量换算276

Antigravity

17.41%
按下载量换算193

OpenCode

12.34%
按下载量换算137

Cursor

7.64%
按下载量换算85

windsurf

3.05%
按下载量换算34

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。

继续浏览同类 Skills