Token导航 LogoToken导航TokenDH.com
开发执行命令github未标认证来源可访问许可证需确认审计通过

tech-learning-guide技术学习指南

Agent Skill

tech-learning-guide 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

870

周安装

37

GitHub Stars

28

下载量

305
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

2

许可证

unknown

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:tech-learning-guide(技术学习指南)
来源仓库:https://github.com/sivaprasadreddy/learning-skill
仓库路径:skills/tech-learning-guide
安装命令:
npx skills add https://github.com/sivaprasadreddy/learning-skill --skill tech-learning-guide
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/sivaprasadreddy/learning-skill --skill tech-learning-guide

简介

tech-learning-guide 用于记录任务执行中的错误、用户纠正和经验缺口。

  • 适合让 Agent 持续沉淀问题、修正和最佳实践。
  • 通过 npx skills add 命令从指定 GitHub 仓库安装使用。
  • 安装前建议确认权限范围和维护状态,避免触发文件读写操作。
  • 适用宿主包括 Codex、Claude、Cursor、Gemini CLI,接入前应确认版本、权限和运行环境要求。

SKILL.md

Tech Learning Guide Skill

Generate a complete, structured, opinionated learning guide for any technical topic. The guide should feel like it was written by an experienced senior engineer who has taught the topic before — practical, direct, and honest about what matters most.

Output Structure

Produce the guide as a rendered artifact (.md file) with the following sections in order. The tone should be warm, direct, and technically confident — avoid fluff and padding.


Section 1 — Overview & Concept Map

Open with 2–3 sentences describing what the technology is and why it matters in today's ecosystem. Then list all key concepts the learner needs to master, grouped into three tiers:

TierLabelWhat it means
🟢BasicsConcepts a beginner must learn first; no prior knowledge of the tech needed
🔵FoundationalCore concepts that form day-to-day working knowledge
🔴AdvancedDeep internals, production concerns, performance, or ecosystem integrations

Format as a table or grouped list. Aim for 5–8 concepts per tier (15–24 total), but adjust naturally for the scope of the technology. Don't pad artificially.


Section 2 — Weekly Learning Schedule

Create a realistic week-by-week study plan. Assume 1–2 hours per day, 5 days/week. Structure:

  • Week N: [Theme] — 2 sentence description of the week's focus

- Day 1–5: specific daily objectives (1–2 sentences each)

Cover all three tiers across the schedule. Basics first, then Foundational, then Advanced. Typical schedule length: 6–10 weeks depending on technology breadth. Be honest if some weeks require more time (e.g., "this week is dense — budget extra time if needed").


Section 3 — Local Development Setup

Provide concrete, copy-paste-ready instructions for getting a working local environment. Include:

  • Prerequisites (OS, required tools, versions)
  • Installation steps (numbered, shell commands in code blocks)
  • Verify it works — a minimal "hello world" command or snippet to confirm setup
  • Recommended IDE / editor with key plugins or extensions
  • Useful dev tools (linters, formatters, debuggers, local test runners)

Be opinionated. If one approach is clearly better for learning, say so.


Section 4 — Concept Deep-Dives

For every concept listed in Section 1, provide a dedicated subsection:

### [Tier emoji] [Concept Name]

**What it is**: 2–3 sentence plain-language explanation.

**Why it matters**: 1–2 sentences on practical relevance.

**How it works**: Explanation with code example(s). For code-heavy concepts, show at least
one realistic example (not just "hello world"). For architectural concepts, use a short
diagram or analogy.

**Common pitfalls**: 1–3 bullet points of mistakes beginners make.

---
#### ✏️ Exercises

1. [Concrete, hands-on exercise — beginner friendly]
2. [Slightly harder exercise that builds on the first]
3. [Challenge exercise that combines this concept with a previous one]

Group the concepts under their tier heading: ## 🟢 Basics, ## 🔵 Foundational, ## 🔴 Advanced


Section 5 — Popular Libraries & Ecosystem

Give a concise ecosystem map. For each major library or tool in the technology's orbit:

  • Name + one-liner on what it does
  • When to use it (2 sentences)
  • A code snippet showing the simplest useful example (where applicable)

Aim for 6–10 entries. Group by category if the ecosystem is large (e.g., "Testing", "HTTP", "ORM"). Don't list everything — list what a practitioner actually reaches for.


Section 6 — Project Ideas

Provide exactly 3 project ideas, progressing in complexity:

  1. [Small] Project Name — 1–2 weeks. What you'll build, which concepts it exercises, why it's a good learning vehicle.
  2. [Medium] Project Name — 3–5 weeks. Same format. Introduce more Foundational concepts.
  3. [Stretch] Project Name — Open-ended. Brings in Advanced concepts and real-world concerns (auth, persistence, deployment, performance).

For each project, include:

  • A brief feature list (3–5 bullet points)
  • Key concepts practiced
  • Optional: suggested tech stack to pair with

Section 7 — What's Next

After mastering everything in this guide, where does the learner go? Provide:

  • 2–3 adjacent technologies to explore (with 1-sentence rationale each)
  • 2–3 advanced topics within the technology itself that weren't covered
  • 1–2 community resources to stay current (conference, newsletter, forum)

Frame this as an honest "graduation" — you've learned the fundamentals; here's the horizon.


Section 8 — Resources

Curated reference list. Organize into categories:

  • Official Docs — link + 1-sentence description
  • Books — title, author, why it's recommended
  • Online Courses — platform + course name, ideal for which learner level
  • Blogs / Sites — name + what kind of content they publish
  • YouTube / Video — channel or specific series, what makes it worth watching
  • Community — Discord, Slack, forum, subreddit

Keep it tight — 3–5 entries per category max. Every entry should earn its place.


Quality Guidelines

  • Be concrete: No vague advice like "practice a lot." Say what to practice and how.
  • Be honest: If a concept is genuinely hard or a common sticking point, say so.
  • Examples must run: Any code snippet should be minimal but complete enough to actually execute or be understood in isolation.
  • Calibrate depth to the technology: A guide for Bash scripting vs. distributed systems should feel different in depth and breadth.
  • No hallucinated resources: Only include books, courses, and links you're confident exist. If unsure about a specific URL, name the resource without a link rather than guess.
  • Developer Advocate tone: Write as a knowledgeable peer, not a textbook. First-person asides are fine ("I'd start with X before Y because...").

Handling Broad vs. Narrow Topics

  • Very broad topic (e.g., "cloud computing", "web development"): Narrow the scope in the intro. Tell the user what angle you're covering and why. Suggest they ask follow-up guides for adjacent areas.
  • Very narrow topic (e.g., a single library method or niche tool): Scale down accordingly. A 10-week schedule for Array.prototype.reduce is absurd. Use judgment.
  • Version-sensitive technology: Mention the version the guide targets at the top. Flag any areas where older versions differ significantly.

References

  • references/schedule-templates.md — Example schedule patterns for different technology scopes (language, framework, tool, concept). Read if you need calibration on schedule length/density.
  • references/concept-tiers.md — Guidance on how to classify concepts into Basics / Foundational / Advanced for common technology categories.

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

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

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

Codex

36.54%
按下载量换算111

Claude

31.36%
按下载量换算96

Cursor

16.88%
按下载量换算51

Gemini CLI

8.38%
按下载量换算26

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

通过

权限和风险

执行命令

安装流程涉及命令执行,可能通过 npx skills add https://github.com/sivaprasadreddy/learning-skill --skill tech-learning-guide 联网下载 Skill 或依赖。用户安装前应确认命令来源、仓库内容和执行环境。

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