← 返回 Skills 市场
315133264

learn-anything

作者 315133264 · GitHub ↗ · v1.0.0 · MIT-0
cross-platform ✓ 安全检测通过
422
总下载
0
收藏
3
当前安装
1
版本数
在 OpenClaw 中安装
/install learn-anything-in-one-hour
功能描述
Teach users any new skill/knowledge X in ~1 hour using a fixed 4-step workflow optimized for complete beginners, focusing on 80/20 rule for maximum value in...
使用说明 (SKILL.md)

Learn Anything in One Hour Skill

Optimized for absolute beginners, delivers 80% of practical value in 60 minutes using the Pareto principle.

Trigger Conditions

Activate when user:

  • Asks to learn a new skill/topic quickly
  • Says "teach me X in one hour" or "learn X fast"
  • Explicitly requests beginner-friendly crash course on a new topic

Core Workflow (follow strictly in order)

Assumption: User is 100% new to topic X, no prior knowledge.

Step 1: Core Logic & 20% Key Points (15 mins)

  1. Explain the core underlying logic of X in 3-5 simple sentences (no jargon)
  2. List exactly ~20% of the most critical key points/concepts that cover 80% of common real-world usage scenarios for X
  3. Keep explanations plain, avoid advanced terminology unless defined immediately

Step 2: MVP Demo & Hands-On Practice (30 mins)

  1. Provide a clear, step-by-step workable MVP roadmap/demo that the user can follow along
  2. The demo must be runnable end-to-end, and cover all core logic from Step 1
  3. Stay available to answer any questions the user has while they work through the demo
  4. Only proceed to next step after the user confirms their demo is working as expected

Step 3: Error Cheat Sheet & Deep Dive (10 mins)

  1. Create a 2-column table listing the most common errors users run into when completing the demo, plus their simple fixes
  2. Add brief explanations for any core concepts that users commonly get confused about during this stage
  3. Make the sheet easy to reference later

Step 4: Summary & Advanced Roadmap (5 mins)

  1. Summarize what the user learned, and confirm they understand the core value of X
  2. Provide a curated advanced study roadmap with 3-5 clear next steps if the user wants to go deeper into X
  3. Link to high-quality, beginner-friendly resources for further learning

Reference Materials

  • references/80-20-topics.md: Pre-curated 20% key point lists for common topics
  • references/common-error-sheets.md: Pre-built error cheat sheets for popular skills
  • references/advanced-roadmap-templates.md: Templates for next step recommendations
安全使用建议
This skill appears coherent and self-contained: it provides a structured 4-step teaching workflow and only uses the included reference files. Before executing any demo steps the agent suggests, manually review commands or links it provides (especially any that download or run third-party code, require elevated privileges, or request credentials). If the agent later asks for environment variables, API keys, or to install software, treat that as a red flag and pause to verify why those are needed. If you want extra assurance, ask the publisher for sample demo steps for the specific topic you plan to learn and review them before running anything.
功能分析
Type: OpenClaw Skill Name: learn-anything-in-one-hour Version: 1.0.0 The skill bundle is a purely educational framework designed to guide an AI agent in teaching new topics using the 80/20 rule. It contains no executable code, no network requests, and no instructions that would lead to data exfiltration or system compromise. All files, including SKILL.md and the reference materials in the references/ directory, consist of standard pedagogical instructions and troubleshooting guides for common beginner topics like Python and Git.
能力评估
Purpose & Capability
Name/description (teach a topic in ~1 hour) match the provided SKILL.md and local reference files. There are no unrelated env vars, binaries, or config paths requested.
Instruction Scope
The SKILL.md strictly defines a 4-step teaching workflow and only references local reference files included in the package. One operational note: Step 2 requires providing a runnable end-to-end demo the user can follow — that could legitimately involve commands, code snippets, or downloading third-party resources in some topics. The skill itself does not include those external downloads or instruct the agent to read system files or secrets, but users should review any demo steps before executing external commands or downloads.
Install Mechanism
No install spec and no code files to execute — instruction-only skill, which is the lowest-risk install mechanism.
Credentials
The skill declares no required environment variables, credentials, or config paths. The packaged reference files are local and predictable.
Persistence & Privilege
always is false, the skill does not request persistent or elevated agent-level privileges, and it does not modify other skills or system-wide settings.
如何使用
  1. 确保已安装 OpenClaw(本地或 Docker 部署)
  2. 在对话框中输入安装命令:/install learn-anything-in-one-hour
  3. 安装完成后,直接呼叫该 Skill 的名称或使用 /learn-anything-in-one-hour 触发
  4. 根据 Skill 的参数说明提供必要输入,即可获得结构化输出
版本历史
v1.0.0
Initial release of Learn Anything in One Hour skill. - Guides users to learn any new skill in about 60 minutes using an optimized 4-step process for beginners. - Focuses on teaching the essential 20% of concepts that deliver 80% of real-world value. - Includes hands-on demo, a clear error cheat sheet, and practical next steps for advanced learning. - Activates when users request a fast, beginner-friendly crash course in a new skill or topic. - Designed for users with zero prior knowledge.
元数据
Slug learn-anything-in-one-hour
版本 1.0.0
许可证 MIT-0
累计安装 3
当前安装数 3
历史版本数 1
常见问题

learn-anything 是什么?

Teach users any new skill/knowledge X in ~1 hour using a fixed 4-step workflow optimized for complete beginners, focusing on 80/20 rule for maximum value in... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 422 次。

如何安装 learn-anything?

在 OpenClaw 或 Claude Code 对话框中运行命令「/install learn-anything-in-one-hour」即可一键安装,无需额外配置。

learn-anything 是免费的吗?

是的,learn-anything 完全免费,采用 MIT-0 许可证,可自由下载、安装和使用。

learn-anything 支持哪些平台?

learn-anything 跨平台运行,可在任意部署了 OpenClaw / Claude Code 的环境中使用(cross-platform)。

谁开发了 learn-anything?

由 315133264(@315133264)开发并维护,当前版本 v1.0.0。

💬 留言讨论