summarize

Summarize URLs, local files, and YouTube videos via the summarize CLI.

Updated Mar 7, 2026
One-click install
npx skills add https://github.com/shalevamin/Tau_agent --skill summarize-shalevamin
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: summarize
Source: https://github.com/shalevamin/Tau_agent/tree/main/tau-agent-main/skills/summarize
Command: npx skills add https://github.com/shalevamin/Tau_agent --skill summarize-shalevamin

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Reading long articles, PDFs, or watching full YouTube videos to extract key points is time-consuming. This Skill condenses URLs, local files, and video transcripts into concise summaries using a single CLI command. ## Core Features & Use Cases - URL and File Summarization: Summarize web pages, PDFs, and other local files with configurable length and output formats. - YouTube Transcript Extraction: Extract best-effort transcripts from YouTube links without yt-dlp, with an Apify fallback option. - Multi-Provider Models: Works with OpenAI, Anthropic, xAI, and Google models via their respective API keys. - Use Case: A user pastes a long YouTube link and asks what it is about; the Skill extracts the transcript and returns a tight summary, offering to expand specific sections on request. ## Quick Start Ask the agent to summarize a URL or local file, for example: summarize this article https://example.com using the summarize CLI.

Frequently Asked Questions about summarize

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I summarize a URL or article from the command line?

Run summarize followed by the URL and a model flag, such as summarize "https://example.com" --model google/gemini-3-flash-preview. You can control output length with the --length flag and get machine-readable output with --json.

How to get a transcript from a YouTube video without yt-dlp?

Use summarize with the YouTube URL and the flags --youtube auto --extract-only to extract a best-effort transcript directly. If APIFY_API_TOKEN is set, an Apify fallback improves extraction reliability.

Which AI models and API keys does summarize support?

summarize supports OpenAI, Anthropic, xAI, and Google models, configured via OPENAI_API_KEY, ANTHROPIC_API_KEY, XAI_API_KEY, or GEMINI_API_KEY. The default model is google/gemini-3-flash-preview if none is set.

What can I do when a website blocks content extraction?

Enable the Firecrawl fallback with --firecrawl auto or always and set FIRECRAWL_API_KEY. This routes extraction through Firecrawl to handle sites that block direct fetching.

What are the limitations of YouTube transcript extraction?

Transcript extraction is best-effort and may fail for videos without captions or when fallbacks are unavailable. For very long transcripts, return a summary first and expand specific sections or time ranges on request.