Document and Web Summarizer
Paste the following prompt into your AI chat to install this skill:
Follow https://skillhub.cn/install/skillhub.md to install @user_7eecc4ea/summarize-new into your AI assistant.
About this skill
Problem
Long documents, web pages, and PDF files often include navigation, ads, tables of contents, and formatting noise, which can distract summarization or hide the main points. Document Summarize targets this by first extracting readable text and then producing a structured summary using fixed analysis dimensions.
How it works
- Web pages: fetch HTML with
WebFetch, then clean non-content markup usingscripts/clean_html.py. - Files: use
scripts/extract_pdf.pyforPDFfiles, readMDandTXTfiles directly, and analyze pasted text or code snippets from the conversation. - Analysis dimensions: identify the topic, section structure, key entities, core arguments, data or facts, and action items.
- Output: generate an executive summary, section highlights, conclusions, and recommended sections for long content; compress short content into a one-line overview and key points.
Boundaries
The skill depends on accessible text and cannot process image-only pages, video, or login-protected content. PDF extraction may require pypdf; if extraction fails, convert the file to text or Markdown. Summaries remain faithful to the source and avoid external inference; very short input is returned with a note that summarization is unnecessary.
Use Cases
- After receiving a competitor’s launch page, extract the main text, key claims, and data for review notes.
- Convert a client contract PDF to text, then summarize clauses, amounts, dates, and action items.
- Paste a long technical blog post or paper excerpt and ask for a readable summary by section, entities, and conclusions.
- Inspect a URL by extracting only readable body text and removing navigation noise before deciding whether to store it.
Best For
- Content editors assembling web material: they need body text, key claims, and quotable data from news pages.
- Legal assistants handling contracts and reports: they need PDF clauses, amounts, dates, and action items summarized for review.
- Engineers tracking technical resources: they want long articles, papers, or API docs compressed into section points and key entities.
- Platform operators managing knowledge bases: they need to judge whether web body text is usable and extract topics, facts, and action items.
Related Skills
Updates Next.js documentation based on code changes in the active branch to help maintainers review PR documentation completeness.
Use a book title and author to search reviews, contents, and reader feedback, then generate a structured breakdown of historical or decision-logic books with key claims, concepts, quotes, and reading notes in Obsidian.
A multi-system divination advisor covering Bazi, Ziwei, tarot, feng shui, timing, and naming with structured interpretations.
Provides Bazi, Liu Yao, Qimen, Meihua, and date-selection guidance from classical references, with plain-language explanations and actionable suggestions.