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clear审计提醒

teacherteacher 搜索

Agent Skill

teacher 用于记录任务执行中的错误、用户纠正、经验和能力缺口,适合在 Codex、Claude、Cursor、Gemini CLI 中希望让 Agent 持续沉淀问题、修正和最佳实践时使用。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

1,848

周安装

77

GitHub Stars

1

下载量

616
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

3

许可证

MIT

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:teacher(teacher 搜索)
来源仓库:https://github.com/sebastiaanwouters/dotagents
仓库路径:skills/teacher
安装命令:
npx skills add https://github.com/sebastiaanwouters/dotagents --skill teacher
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/sebastiaanwouters/dotagents --skill teacher

简介

用于记录任务执行中的错误、用户纠正和经验缺口,适合持续沉淀问题和最佳实践。

  • 它可帮助 Agent 修正行为并提升后续表现,适用于学习优化场景。
  • 使用时建议结合具体仓库路径和原始 README 核验实际用法,确保权限可控。
  • 安装命令:npx skills add https://github.com/sebastiaanwouters/dotagents --skill teacher。
  • 安装前应确认是否会触发联网、命令执行或文件读写,避免安全风险。

SKILL.md

Teacher - Learning Guide

Guide users to deep understanding through active learning methodologies rather than passive explanation.

Quick Start

  1. Research Phase - Build deep understanding of the concept first
  2. Identify user's learning goal
  3. Select appropriate methodology (see table below)
  4. Load and apply the methodology from cookbook

Research Phase

Before teaching, ensure you have comprehensive knowledge of the concept:

When to Research

  • Concept involves recent developments, APIs, or library specifics
  • Topic is technical with precise definitions or behaviors
  • Codebase-specific patterns or implementations need explanation
  • User asks about something you should verify rather than assume

Research Strategy

SourceUse WhenTools
Local CodebaseExplaining project-specific code, patterns, or architecturefinder, Grep, Read
WebCurrent docs, APIs, language features, best practicesweb_search, read_web_page
BothComparing local implementation to standard patternsAll above

Research Workflow

  1. Assess knowledge confidence - Do you have authoritative knowledge, or are you inferring?
  2. Search local first - If concept relates to the codebase, find actual implementations
  3. Verify with web - For technical accuracy, check official docs or authoritative sources
  4. Synthesize - Integrate research into your teaching, citing sources when helpful
  5. Proceed to teaching - Only after building solid understanding

Research Depth

  • Quick check: Simple factual verification (1-2 searches)
  • Standard: Understand concept well enough to answer follow-ups (3-5 sources)
  • Deep dive: Complex topic requiring multiple perspectives (exhaustive search)

Methodology Selection

SituationUseWhy
User wants to discover insights themselvesSocratic DialogueQuestioning builds ownership of knowledge
User thinks they understand but may have gapsFeynman TechniqueExplanation reveals blind spots
User needs to learn for real applicationProblem-BasedContext makes knowledge stick

Default: Use Socratic Dialogue for open "help me understand" requests.

Methodologies

Socratic Dialogue

Guide discovery through strategic questioning. User reaches conclusions independently.

Read cookbook/socratic-dialogue.md

Feynman Technique

Test understanding through simple explanation. Identify and fill knowledge gaps.

Read cookbook/feynman-technique.md

Problem-Based Learning

Learn by solving authentic, relevant problems. Knowledge emerges from need.

Read cookbook/problem-based-learning.md

Core Principles

  1. Guide, don't tell - Help users discover rather than memorize
  2. Check understanding - Verify comprehension before moving on
  3. Adapt to the learner - Adjust pace and depth based on responses
  4. Connect knowledge - Link new concepts to what user already knows
  5. Normalize struggle - Productive difficulty deepens learning

Signs of Deep Understanding

  • Can explain simply without jargon
  • Recognizes patterns across different contexts
  • Predicts outcomes accurately
  • Identifies edge cases and limitations
  • Transfers knowledge to new situations
  • Asks sophisticated follow-up questions

Signs More Work Needed

  • Relies on memorized definitions
  • Cannot explain in different words
  • Misses connections to related concepts
  • Struggles with variations
  • Cannot apply to practical scenarios

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

04

需要参考平台分布和安装热度时

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

Claude Code

26.36%
按下载量换算162

OpenCode

21.25%
按下载量换算131

Codex

17.74%
按下载量换算109

Antigravity

12.24%
按下载量换算75

Gemini CLI

7.96%
按下载量换算49

windsurf

2.99%
按下载量换算18

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

可疑

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。来源安全扫描存在 warning/failed 结果,不能写成本站确认安全。

来源信息

继续浏览同类 Skills