How do I install the Summarize skill?

Run npx skillmds@latest add mj-deving/summarize 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 Summarize safe to use?

This skill has not completed SkillMD's automated safety review yet. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Summarize?

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 Summarize free to use?

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

Who published Summarize?

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

Summarize by mj-deving — Agent Skill

Name: Summarize
Author: mj-deving

Summarize

Universal content extraction, summarization, web scraping, and transcription via summarize CLI (v0.11.1). USE WHEN summarize URL, extract transcript, YouTube transcript, summarize video, summarize article, batch summarize, extract slides, transcribe audio, summarize webpage, scrape website, extract markdown, web extraction, content extraction, scrape URL, podcast transcript, audio to text, video slides, OCR slides, extract content from URL.

mj-deving a663da8 7 files · 25.1 KB Updated Sep 17, 2026

File contents

Customization

Before executing, check for user customizations at: ${PAI_USER_DIR}/SKILLCUSTOMIZATIONS/Summarize/

If this directory exists, load and apply any PREFERENCES.md, configurations, or resources found there. These override default behavior.

Summarize — Universal Content Extraction & Summarization

Five capabilities in one CLI tool: Summarize, Scrape, Transcribe, Extract Slides, Batch Process.

Workflow Routing

Request Pattern	Route To
Summarize URL/video/article, key takeaways from URL	`Workflows/SummarizeContent.md`
Extract transcript, YouTube transcript, transcribe audio/video	`Workflows/ExtractTranscript.md`
Scrape URL, extract webpage, get markdown from site	`Workflows/ScrapeWebpage.md`
Summarize multiple URLs, batch process	`Workflows/BatchProcess.md`
Extract slides, slides from video, OCR slides	`Workflows/SlideExtract.md`

Quick Decision Tree

User wants...
├── Summary of content → SummarizeContent
├── Raw text/transcript → ExtractTranscript
├── Webpage as Markdown → ScrapeWebpage
├── Multiple URLs processed → BatchProcess
└── Visual slides from video → SlideExtract

Environment Setup

Source API keys before any summarize call:

[ -f ${AGENT_HOME}/.env ] && export $(grep -v '^#' ${AGENT_HOME}/.env | xargs)
export YT_DLP_PATH=$(which yt-dlp 2>/dev/null || echo ${HOME}/.local/bin/yt-dlp)

Integration with ExtractWisdom

Summarize extracts and compresses. ExtractWisdom analyzes in depth. Pipeline: summarize --extract → raw text → ExtractWisdom → structured insights.

Scraping Tier Model

Tier 0: WebFetch (built-in, fastest, limited to static HTML)
Tier 1: Summarize --extract --format md (Readability + optional Firecrawl)
Tier 2: Scraping-Skill Bright Data Proxy (anti-bot bypass)
Tier 3: Scraping-Skill Playwright (full browser rendering)
Tier 4: Scraping-Skill Apify Actors (social media platforms)

Use Summarize for Tier 1 before escalating to heavier tools.

---name: summarizedescription: Universal content extraction, summarization, web scraping, and transcription via summarize CLI (v0.11.1). USE WHEN summarize URL, extract transcript, YouTube transcript, summarize video, summarize article, batch summarize, extract slides, transcribe audio, summarize webpage, scrape website, extract markdown, web extraction, content extraction, scrape URL, podcast transcript, audio to text, video slides, OCR slides, extract content from URL.---## Customization**Before executing, check for user customizations at:**`${PAI_USER_DIR}/SKILLCUSTOMIZATIONS/Summarize/`If this directory exists, load and apply any PREFERENCES.md, configurations, or resources found there. These override default behavior.# Summarize — Universal Content Extraction & SummarizationFive capabilities in one CLI tool: **Summarize**, **Scrape**, **Transcribe**, **Extract Slides**, **Batch Process**.## Workflow Routing| Request Pattern | Route To ||---|---|| Summarize URL/video/article, key takeaways from URL | `Workflows/SummarizeContent.md` || Extract transcript, YouTube transcript, transcribe audio/video | `Workflows/ExtractTranscript.md` || Scrape URL, extract webpage, get markdown from site | `Workflows/ScrapeWebpage.md` || Summarize multiple URLs, batch process | `Workflows/BatchProcess.md` || Extract slides, slides from video, OCR slides | `Workflows/SlideExtract.md` |## Quick Decision Tree```User wants...├── Summary of content → SummarizeContent├── Raw text/transcript → ExtractTranscript├── Webpage as Markdown → ScrapeWebpage├── Multiple URLs processed → BatchProcess└── Visual slides from video → SlideExtract```## Environment SetupSource API keys before any summarize call:```bash[ -f ${AGENT_HOME}/.env ] && export $(grep -v '^#' ${AGENT_HOME}/.env | xargs)export YT_DLP_PATH=$(which yt-dlp 2>/dev/null || echo ${HOME}/.local/bin/yt-dlp)```## Integration with ExtractWisdomSummarize extracts and compresses. ExtractWisdom analyzes in depth.Pipeline: `summarize --extract` → raw text → ExtractWisdom → structured insights.## Scraping Tier Model```Tier 0: WebFetch (built-in, fastest, limited to static HTML)Tier 1: Summarize --extract --format md (Readability + optional Firecrawl)Tier 2: Scraping-Skill Bright Data Proxy (anti-bot bypass)Tier 3: Scraping-Skill Playwright (full browser rendering)Tier 4: Scraping-Skill Apify Actors (social media platforms)```Use Summarize for Tier 1 before escalating to heavier tools.

Frequently asked questions