summarize-all

Summarize web pages, PDFs, images, audio, and YouTube videos via OpenAI-compatible APIs.

1|Updated Jul 13, 2026
One-click install
npx skills add https://github.com/HalseyYang/Skills-HY --skill summarize-all-halseyyang
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: summarize-all
Source: https://github.com/HalseyYang/Skills-HY/tree/main/summarize-all
Command: npx skills add https://github.com/HalseyYang/Skills-HY --skill summarize-all-halseyyang

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve? Reading long articles, papers, videos, and audio files takes significant time, and manually tracking, comparing, and organizing summaries across sources is tedious. This Skill automates content summarization with caching, tagging, and history search. ## Core Features & Use Cases - Multi-Source Summarization: Summarize web pages, PDFs, images, audio files, and YouTube videos with automatic content-type detection and adaptive summary length. - Advanced Modes: Q&A on content, multi-article comparison, academic paper analysis, translation into 7 languages, knowledge graphs, and structured reports. - Organization & Automation: 24-hour caching, automatic tag extraction, keyword tracking alerts, history search, batch processing, and an HTTP API server with webhook support. - Use Case: A researcher collects 10 URLs on a topic, runs batch mode to summarize them all, then uses compare mode to identify common themes and contradictions across sources. ## Quick Start Configure your API endpoint and key, then ask the assistant to summarize a URL or local file, for example: summarize this article https://example.com with key points and tags.

Frequently Asked Questions about summarize-all

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

FAQPage Schema
How do I summarize a web page or PDF from the command line?

Run summarize-all followed by the URL or file path, such as summarize-all "https://example.com" or summarize-all "/path/file.pdf". The tool fetches the content, detects its type, and returns a summary with key points and tags.

What content types can this summarizer handle?

It supports web pages, PDFs, images (jpg, png, gif, webp), audio files (mp3, wav, m4a, ogg), YouTube videos, and plain text. PDF extraction requires pdftotext, and YouTube or audio transcription requires a summarize CLI or yt-dlp.

Which AI models and API providers are supported?

Any OpenAI-compatible chat completions endpoint works, including GPT-4o, GPT-4o-mini, Claude or Gemini via proxy, OpenRouter models, and Chinese models like Qwen, DeepSeek, Kimi, GLM, and MiniMax. You configure the endpoint, key, and model yourself.

Does the summarizer cache results to avoid repeated API calls?

Yes, summaries are cached for 24 hours in ~/.config/summarize-all/cache.json. Repeated requests for the same URL and length return the cached result unless you pass the --no-cache flag.

Can I compare multiple articles or ask questions about content?

Yes, use summarize-all compare with two or more URLs to get common themes, differences, and credibility ranking, or use summarize-all qa with a URL and a question to query specific content.

Is my content private when using this summarization tool?

All content you summarize is sent to the API endpoint you configure. You control which endpoint, key, and model process your data, so use a trusted provider that meets your privacy requirements.