This turns articles, papers, or books into sequences of sharp questions with structured answers. Instead of summarizing, it extracts the author's core argument as a chain of "why does this work / what's the tradeoff / where does it break" questions. Each answer follows a strict format: one-line conclusion, formalized relationship (like "old: X → new: Y"), reasoning steps, and boundary conditions. It's designed for intellectual reconstruction, not FAQ generation. The questions are supposed to bite into the real substance, and answers are meant to drive stakes into the ground so you can retrace the original logic. Works with URLs, PDFs, or raw text. Outputs to org-mode files.

skills.sh installs
4.9K
GitHub stars
7.7K
last updated
Aug 21, 2026

Install

$ npx -y skills add lijigang/ljg-skills --skill ljg-qa --agent claude-code

Installs into .claude/skills of the current project.

Files

ljg-qa: 问答提取

读一份东西,把它的思想拆成「为什么—怎么—边界」的问答链。

读者顺着 Q 走过去,每个 A 砸下来一枚钉子。

你不是

  • 不是 FAQ 生成器("什么是 X"——读者一看就跳过)
  • 不是摘要换皮(把段落拆成"问/答"两半还是摘要)
  • 不是知识点列表(孤立的事实碰撞不出洞察)
  • 不是阅读理解题(提问不是为了考读者,是为了切中作者)

你是

把作者的论证骨架翻出来,每根骨头长成一个尖锐的问题。读者沿着 Q 链读,能复现作者的整套思路——而不是被告知结论。

三条铁律

  1. Q 切要害 —— 问的是「为什么这个解法成立」「它跟另一种做法差在哪」「它的代价是什么」「它在哪里失效」,不是「它定义是什么」。一个 Q 必须能让答案承重,不能被一句话敷衍过去。

  2. A 有形式化收口 —— 每个 A 严格四段:结论(一句话)+ 形式化(用文字 + 简单符号把思想压成一行可视关系,如 A = B + C、旧: X → 新: Y)+ 论证步(怎么想到的)+ 边界(不成立的条件)。形式化是"思想的几何",让读者一眼看出关系。

  3. Q 链有方向 —— Q 之间不是并列罗列,是「Q1 答完→Q2 自然冒出来」。读者读完整串 Q,相当于走了一遍作者的推理路径。

工作流

按 Workflows/Extract.md 的步骤执行。

设计参考

Q 怎么提、A 怎么收口的具体模式见 References/QuestionDesign.md。

Voice Notification

执行 workflow 时:

curl -s -X POST http://localhost:31337/notify \
  -H "Content-Type: application/json" \
  -d '{"message": "Running Extract in ljg-qa"}' \
  > /dev/null 2>&1 &

输出文本:

Running **Extract** in **ljg-qa**...

输出

  • 格式:Markdown(**bold**,禁 Org 语法)
  • 路径:~/Documents/notes/
  • denote 文件名:{YYYYMMDDTHHMMSS}--qa-{核心主题 5-10 字}__qa.md

Examples

Example 1: URL

Sponsors

Details

Repository
github.com/lijigang/ljg-skills
Categories
Testing & QA, Sales & Marketing
First seen
May 16, 2026

More from lijigang/ljg-skills

All 25
1agricidaniel avatarads-testagricidaniel/claude-adsDesign and evaluate paid-ad experiments with hypotheses, randomization units, sample-size and duration assumptions, guardrails, platform experiment tools, analy3.9Kskills.sh installs
2microsoft avatarplaywright-devmicrosoft/playwrightExplains how to develop Playwright - add APIs, MCP tools, CLI commands, and vendor dependencies.3.8Kskills.sh installs
3pytorch avatarpyrefly-type-coveragepytorch/pytorchMigrate a file to use stricter Pyrefly type checking with annotations required for all functions, classes, and attributes.3.2Kskills.sh installs
4phuryn avatarab-test-analysisphuryn/pm-skillsAnalyze A/B test results with statistical significance, sample size validation, confidence intervals, and ship/extend/stop recommendations. Use when evaluating 3Kskills.sh installs
5eronred avatarab-test-store-listingeronred/aso-skillsWhen the user wants to A/B test App Store product page elements to improve conversion rate. Also use when the user mentions "A/B test", "product page optimizati2.5Kskills.sh installs
6owl-listener avatardesign-qa-checklistowl-listener/designer-skillsCreate QA checklists for verifying design implementation accuracy.2.5Kskills.sh installs