content-summary

name	content-summary
description	Summarize and distill existing content from pasted text, chat context, or already parsed attachment content into summaries, TL;DR, key points, action items, risks, conclusions, decisions, meeting notes, or requirements summaries. Use when the user asks for 总结, 摘要, 提炼重点, 归纳, 归纳需求点, 行动项, 风险, 结论, 会议纪要, summarize, summary, TL;DR, key points, action items, risks, conclusions, decisions, or meeting notes.
when_to_use	Use this skill only after the source content is already available in the conversation, pasted by the user, or supplied by the platform attachment parser. Do not use it to read, parse, extract, OCR, or inspect uploaded PDF, DOCX, XLSX, CSV, image, or other files directly. If the user asks to export the summary as Word, PDF, Markdown, or another file, produce the summary first and then route the generated content to file-generator.
runtime_type	prompt
max_calls_per_turn	5
timeout_seconds	5
display	{"icon":"list-collapse","category":"document_processing","scenarios":["document_handling","office_collaboration","knowledge_research"],"label":{"en_US":"Content Summary","zh_Hans":"内容总结"},"description":{"en_US":"Designed for long text, meeting notes, or parsed documents that need faster reading; extracts summaries, key points, actions, risks, conclusions, and meeting minutes.","zh_Hans":"适用于快速理解长文本、会议记录或已解析文档，可提炼摘要、重点、行动项、风险、结论和会议纪要。"},"when_to_use":{"en_US":"Use when existing text or extracted document content needs to be summarized or distilled.","zh_Hans":"当已有文本或文档读取结果需要总结、提炼或归纳时使用。"},"tags":{"en_US":["Summary","Extraction","Productivity"],"zh_Hans":["总结","提炼","归纳"]}}

Content Summary Skill

Use this skill to compress, distill, and structure existing content. This skill does not read files, parse file formats, OCR images, generate files, or create downstream artifacts.

Scope

Use this skill for:

• Summaries, abstracts, TL;DR, executive summaries, and concise overviews.
• Key points, important facts, decisions, conclusions, assumptions, and open questions.
• Action items, owners, deadlines, dependencies, blockers, and follow-ups.
• Risks, issues, mitigations, tradeoffs, and unresolved concerns.
• Meeting notes, discussion summaries, decisions, and next steps.
• Requirements summaries from already extracted requirement source content.

Do not use this skill to:

• Read, parse, extract, or inspect uploaded files directly.
• Process PDF, DOCX, XLSX, CSV, image, or scanned document bytes.
• Generate Word, PDF, Markdown, TXT, CSV, XLSX, or any downloadable file.
• Invent source facts, owners, dates, decisions, requirements, or risks not present in the source content.

Routing Rules

• If the user directly pastes text and asks to summarize, use this skill only.
• If the user uploads a file and asks to read, extract, parse, inspect, or preserve original content, rely on the platform's parsed attachment content; do not use this skill unless a summary or distillation is requested.
• If the user uploads a file and asks to summarize, distill, or extract conclusions, use this skill only after the attachment parser has supplied source content in the conversation.
• If the user says "read it and summarize", summarize the parsed attachment content that is already available in the conversation.
• If the user asks to export the summary as Word, PDF, Markdown, or another file, first produce the summary text, then route that text to file-generator.

Workflow

1. Identify the source content. It must be pasted text, prior chat context, parsed attachment content, or another extraction result already available in the conversation.
2. Identify the requested output type: general summary, TL;DR, key points, action items, risks, conclusions, meeting notes, or requirements summary.
3. If the output type is clear, read exactly one relevant reference before writing the summary.
4. Preserve important qualifiers, numbers, dates, owners, decisions, and constraints from the source. Mark uncertain or missing details instead of inventing them.
5. Choose a concise Markdown format unless the user requests another format.
6. If the source is long, prioritize user-requested aspects first, then include only high-signal supporting details.