How do I install the My Defuddle skill?

Run npx skillmds@latest add jerryyrliu-jpg/my-defuddle in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is My Defuddle safe to use?

This skill has not completed SkillMD's automated safety review yet. Capability flags: makes network calls, reads secrets. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with My Defuddle?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is My Defuddle free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published My Defuddle?

jerryyrliu-jpg (@jerryyrliu-jpg) published this skill. Their other Agent Skills are listed on their SkillMD profile.

My Defuddle by jerryyrliu-jpg — Agent Skill

Name: My Defuddle
Rating: 4.2 (5 reviews)
Author: jerryyrliu-jpg

File contents

My-Defuddle: Ultimate Content Extractor

This skill is a superset of the original defuddle skill, combining high-efficiency web parsing with advanced AI post-processing for YouTube and Obsidian.

Description

Defuddle is designed to streamline web content ingestion. It strips away navigation menus, advertisements, and clutter, providing a lean Markdown version of articles and documentation to optimize token usage.

Triggers

/my-defuddle <URL>
"extract article from "
"clean this page: "
"Analyze this YouTube video: "

Usage & Mental Model

1. General Web Pages (Original Defuddle Logic)

Defuddle is a server-side HTML parser—it downloads HTML, strips boilerplate, and returns clean prose. It is faster and cheaper than browser-based fetch for standard pages.

When to apply:

Standard HTTP pages (documentation, articles, blog posts, wikis).

When NOT to apply:

Single Page Applications (SPAs) requiring JavaScript execution.
Login-gated pages or pages behind authentication.
JSON API endpoints.

2. YouTube Links (Enhanced AI Layer)

When the URL is from youtube.com or youtu.be:

Raw Extraction: Use defuddle parse <URL> --markdown to get the transcript.
Exhaustive Deep Analysis: Do not omit any core tips, steps, or unique insights.
- 核心主旨 (Core Mission): Use > [!abstract] for the primary goal.
- 內容深度詳解 (Deep Breakdown): Detailed explanation of every technique/chapter. Use nested bullet points.
- 實作建議 (Actionable Advice): Use > [!todo] for concrete steps.
Key Takeaways: Provide 5-10 specific, high-value insights.
Automated Metadata: Generate 5+ #hashtags and append #yyyy-mm-dd.

Output Format (Strict)

# [Title]
## 💎 影片重點分析 (繁體中文) (For YouTube) OR ## 📝 內容深度總結 (For Web)
## 🚀 實作建議 (Actionable Advice)
## 📌 關鍵摘要 (Key Takeaways)
## 📜 原始內容/逐字稿 (Full Content)
- Insert the 100% complete raw text from defuddle here.
## 標籤與日期

Implementation Rules

Always use Traditional Chinese for analysis and summaries.
Ensure the #yyyy-mm-dd tag is correctly formatted.
Data Integrity: Never truncate the raw transcript or main prose in the "Full Content" section.

Based on kepano/obsidian-skills with OpenClaw enhancements.

---name: my-defuddledescription: Ultimate web & YouTube content extractor. Combines Kepano's original defuddle logic with enhanced YouTube analysis, Traditional Chinese summaries, and Obsidian-optimized formatting.---# My-Defuddle: Ultimate Content ExtractorThis skill is a superset of the original `defuddle` skill, combining high-efficiency web parsing with advanced AI post-processing for YouTube and Obsidian.## DescriptionDefuddle is designed to streamline web content ingestion. It strips away navigation menus, advertisements, and clutter, providing a lean Markdown version of articles and documentation to optimize token usage.## Triggers- `/my-defuddle <URL>`- "extract article from <URL>"- "clean this page: <URL>"- "Analyze this YouTube video: <URL>"## Usage & Mental Model### 1. General Web Pages (Original Defuddle Logic)Defuddle is a server-side HTML parser—it downloads HTML, strips boilerplate, and returns clean prose. It is faster and cheaper than browser-based fetch for standard pages.**When to apply:**- Standard HTTP pages (documentation, articles, blog posts, wikis).**When NOT to apply:**- Single Page Applications (SPAs) requiring JavaScript execution.- Login-gated pages or pages behind authentication.- JSON API endpoints.### 2. YouTube Links (Enhanced AI Layer)When the URL is from `youtube.com` or `youtu.be`:1. **Raw Extraction**: Use `defuddle parse <URL> --markdown` to get the transcript.2. **Exhaustive Deep Analysis**: **Do not omit any core tips, steps, or unique insights.** - **核心主旨 (Core Mission)**: Use `> [!abstract]` for the primary goal. - **內容深度詳解 (Deep Breakdown)**: Detailed explanation of every technique/chapter. Use nested bullet points. - **實作建議 (Actionable Advice)**: Use `> [!todo]` for concrete steps.3. **Key Takeaways**: Provide 5-10 specific, high-value insights.4. **Automated Metadata**: Generate 5+ `#hashtags` and append `#yyyy-mm-dd`.## Output Format (Strict)- `# [Title]`- `## 💎 影片重點分析 (繁體中文)` (For YouTube) OR `## 📝 內容深度總結` (For Web)- `## 🚀 實作建議 (Actionable Advice)`- `## 📌 關鍵摘要 (Key Takeaways)`- `## 📜 原始內容/逐字稿 (Full Content)` - *Insert the 100% complete raw text from defuddle here.*- `## 標籤與日期`## Implementation Rules- Always use **Traditional Chinese** for analysis and summaries.- Ensure the `#yyyy-mm-dd` tag is correctly formatted.- **Data Integrity**: Never truncate the raw transcript or main prose in the "Full Content" section.---*Based on [kepano/obsidian-skills](https://github.com/kepano/obsidian-skills) with OpenClaw enhancements.*

Frequently asked questions

My Defuddle