Token导航 LogoToken导航

Agent Skill · 研究检索

tutor-setup(导师设置)

tutor-setup 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。

tutor-setup 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;许可证信息还不明确,第三方扫描有提醒项,按安装量在研究检索类的 73,675 个 Skill 里排第 52,216。

总安装 563周安装 23GitHub Stars 812下载 180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
unknown

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:tutor-setup(导师设置)
来源仓库:https://github.com/bevibing/tutor-skills
仓库路径:skills/tutor-setup
安装命令:
npx skills add https://github.com/bevibing/tutor-skills --skill tutor-setup
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/bevibing/tutor-skills --skill tutor-setup

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。

自动检测项目类型(代码库或文档)并引导 Tutor 系统初始化流程。强制限定操作范围为当前工作目录,防止越权访问外部文件。根据 package.json、Cargo.toml 等项目标记决定进入何种模式。使用前应将外部资料复制至 CWD 后再进行操作。

tutor-setup 属于研究检索类 Skill,可作为该场景下的辅助能力补充。

来源仓库内路径:skills/tutor-setup

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/bevibing/tutor-skills --skill tutor-setup;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

tutor-setup 怎么安装?

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

tutor-setup 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。

tutor-setup 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。

tutor-setup 是免费的吗?

公开资料没有标注明确许可证,收费和商用限制都无法确认。正式使用前请到来源仓库核对授权说明。

tutor-setup 用的人多吗?

公开数据是累计安装 563 次,近一周安装 23 次,下载 180 次,GitHub 812 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。

tutor-setup 安全吗?

第三方扫描有提醒项,说明存在需要人工确认的行为。安装前请打开源码看清它读了什么、联网到哪里。

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。

先想清楚的地方

  • 许可证没有明确标注,能不能商用、能不能再分发都不确定。
  • 第三方扫描存在提醒项或未通过项,需要人工逐条确认。
  • 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。

不太适合

  • 需要明确授权才能用到公司项目里的场景。
  • 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
tutor-setup

导师设置 · 本页

563812需要联网未明确有提醒

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

Tutor Setup — Knowledge to Obsidian StudyVault

CWD Boundary Rule (ALL MODES)

NEVER access files outside the current working directory (CWD). All source scanning, reading, and vault output MUST stay within CWD and its subdirectories. If the user provides an external path, ask them to copy the files into CWD first.

Mode Detection

On invocation, detect mode automatically:

  1. Check for project markers in CWD:

- package.json, pom.xml, build.gradle, Cargo.toml, go.mod, Makefile, *.sln, pyproject.toml, setup.py, Gemfile

  1. If any marker found → Codebase Mode
  2. If no marker found → Document Mode
  3. Tie-break: If .git/ is the sole indicator and no source code files (*.ts, *.py, *.java, *.go, *.rs, etc.) exist, default to Document Mode.
  4. Announce detected mode and ask user to confirm or override.

Document Mode

Transforms knowledge sources (PDF, text, web, epub) into study notes. Templates: templates.md

Phase D1: Source Discovery & Extraction

  1. Auto-scan CWD for **/*.pdf, **/*.txt, **/*.md, **/*.html, **/*.epub (exclude node_modules/, .git/, dist/, build/, StudyVault/). Present for user confirmation.
  2. Extract text (MANDATORY tools):

- PDF → pdftotext CLI ONLY (run via Bash tool). NEVER use the Read tool directly on PDF files — it renders pages as images and wastes 10-50x more tokens. Convert to .txt first, then Read the .txt file. pdftotext "source.pdf" "/tmp/source.txt" - If pdftotext is not installed, install it first: brew install poppler (macOS) or apt-get install poppler-utils (Linux). - URL → WebFetch - Other formats (.md, .txt, .html) → Read directly.

  1. Read extracted .txt files — understand scope, structure, depth. Work exclusively from the converted text, never from the raw PDF.
  2. Source Content Mapping (MANDATORY for multi-file sources):

- Read cover page + TOC + 3+ sample pages from middle/end for EVERY source file - NEVER assume content from filename — file numbering often ≠ chapter numbering - Build verified mapping: {source_file → actual_topics → page_ranges} - Flag non-academic files and missing sources - Present mapping to user for verification before proceeding

Phase D2: Content Analysis

  1. Identify topic hierarchy — sections, chapters, domain divisions.
  2. Separate concept content vs practice questions.
  3. Map dependencies between topics.
  4. Identify key patterns — comparisons, decision trees, formulas.
  5. Full topic checklist (MANDATORY) — every topic/subtopic listed. Drives all subsequent phases.
Equal Depth Rule: Even a briefly mentioned subtopic MUST get a full dedicated note supplemented with textbook-level knowledge.
  1. Classification completeness: When source enumerates categories ("3 types of X"), every member gets a dedicated note. Scan for: "types of", "N가지", "categories", "there are N".
  2. Source-to-note cross-verification (MANDATORY): Record which source file(s) and page range(s) cover each topic. Flag untraceable topics as "source not available".

Phase D3: Tag Standard

Define tag vocabulary before creating notes:

  • Format: English, lowercase, kebab-case (e.g., #data-hazard)
  • Hierarchy: top-level → domain → detail → technique → note-type
  • Registry: Only registered tags allowed. Detail tags co-attach parent domain tag.

Phase D4: Vault Structure

Create StudyVault/ with numbered folders per templates.md. Group 3-5 related concepts per file.

Phase D5: Dashboard Creation

Create 00-Dashboard/: MOC, Quick Reference, Exam Traps. See templates.md.

  • MOC: Topic Map + Practice Notes + Study Tools + Tag Index (with rules) + Weak Areas (with links) + Non-core Topic Policy
  • Quick Reference: every heading includes → [[Concept Note]] link; all key formulas
  • Exam Traps: per-topic trap points in fold callouts, linked to concept notes

Phase D6: Concept Notes

Per templates.md. Key rules:

  • YAML frontmatter: source_pdf, part, keywords (MANDATORY)
  • source_pdf MUST match verified Phase D1 mapping — never guess from filename
  • If unavailable: source_pdf: 원문 미보유
  • [[wiki-links]], callouts ([!tip], [!important], [!warning]), comparison tables > prose
  • ASCII diagrams for processes/flows/sequences
  • Simplification-with-exceptions: general statements must note edge cases

Phase D7: Practice Questions

Per templates.md. Key rules:

  • Every topic folder MUST have a practice file (8+ questions)
  • Active recall: answers use > [!answer]- 정답 보기 fold callout
  • Patterns use > [!hint]- / > [!summary]- fold callouts
  • Question type diversity: ≥60% recall, ≥20% application, ≥2 analysis per file
  • ## Related Concepts with [[wiki-links]]

Phase D8: Interlinking

  1. ## Related Notes on every concept note
  2. MOC links to every concept + practice note
  3. Cross-link concept ↔ practice; siblings reference each other
  4. Quick Reference sections → [[Concept Note]] links
  5. Weak Areas → relevant note + Exam Traps; Exam Traps → concept notes

Phase D9: Self-Review (MANDATORY)

Verify against quality-checklist.md Document Mode section. Fix and re-verify until all checks pass.


Codebase Mode

Generates a new-developer onboarding StudyVault from a source code project. Full workflow: codebase-workflow.md Templates: codebase-templates.md

Phase Summary

PhaseNameKey Action
C1Project ExplorationScan files, detect tech stack, read entry points, map directory layout
C2Architecture AnalysisIdentify patterns, trace request flow, map module boundaries and data flow
C3Tag StandardDefine #arch-*, #module-*, #pattern-*, #api-* tag registry
C4Vault StructureCreate StudyVault/ with Dashboard, Architecture, per-module, DevOps, Exercises folders
C5DashboardMOC (Module Map + API Surface + Getting Started + Onboarding Path) + Quick Reference
C6Module NotesPer-module notes: Purpose, Key Files, Public Interface, Internal Flow, Dependencies
C7Onboarding ExercisesCode reading, configuration, debugging, extension exercises (5+ per major module)
C8InterlinkingCross-link modules, architecture ↔ implementations, exercises ↔ modules
C9Self-ReviewVerify against quality-checklist.md Codebase Mode section

See codebase-workflow.md for detailed per-phase instructions.


Language

  • Match source material language (Korean → Korean notes, etc.)
  • Tags/keywords: ALWAYS English

平台分布

Codex

33.04%
按下载量换算59

Claude

32.31%
按下载量换算58

Cursor

17.51%
按下载量换算32

Gemini CLI

9.76%
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。来源安全扫描存在提醒或未通过项,安装前请自行复核。目前只有一个公开来源,建议同时核对源仓库。

继续浏览同类 Skills