
ljg-qa Idea Extractor
Turns an article, paper or book into a chain of sharp questions and structured answers
- Repository
- lijigang/ljg-skills
- Stars
- 7,607
- Licence
- MIT
- Files
- 3 files
What it does
Once installed, when you share an article, paper or book and ask for Q&A extraction, the agent reads it, finds the author's core reasoning, and rewrites it as a sequence of pointed questions (why this approach works, how it differs from alternatives, its costs and limits). Each answer is a short conclusion, a compact symbolic summary of the relationship, the reasoning step, and the conditions under which it fails. The output is a Markdown note saved to a notes folder, meant to let a reader retrace the author's logic rather than just read a summary.
Great for
- Readers who want to deeply understand the argument of a dense article, paper or book
- Note-takers who keep a Markdown or denote-style notes folder
- Anyone who wants ideas as a logical chain instead of a summary or FAQ
Skip it if
- Study quizzes, FAQs or glossaries: the skill explicitly rejects these formats, so it will not fit exam prep or definition lists
- Quick summaries: the output is longer and more demanding than a simple recap
- Sources you cannot legally or privately send to an agent, since the content is processed by the agent
Try asking
- “/ljg-qa https://example.com/article”
- “/ljg-qa ~/Downloads/paper.pdf”
- “Turn this passage into Q-A pairs: [paste text]”
- “问答:请把这篇文章的核心观点抽成问答链”
Install in 3 steps
- 1
Download the skill
The ZIP contains the 3 original files, the MIT licence and an attribution note.
ljg-qa.zip - 2
Add it to your agent
Claude Code (all projects)
mkdir -p ~/.claude/skills && unzip ljg-qa.zip -d ~/.claude/skills/Claude Code / Agent SDK (this project only)
mkdir -p .claude/skills && unzip ljg-qa.zip -d .claude/skills/Claude apps (web/desktop): Settings → Capabilities → Skills → Upload skill, and choose the ZIP.
- 3
Ask for the task
No special command needed — the agent reads the skill's description and loads it when your request matches. Try one of the prompts above.
Get better results
- Give it a complete article or a specific PDF page range so the question chain has enough material to order by argument dependency
- Review the generated answers against the source, since the symbolic summaries can oversimplify nuanced claims
- Check the notes folder path before running so output lands where you expect
What's inside
- References/QuestionDesign.md
- SKILL.md
- Workflows/Extract.md
Redistributed unmodified from lijigang/ljg-skills under the MIT licence. The assessment was drafted with AI from the skill's files on 2026-10-29. Skills can instruct your agent to run code — read SKILL.md and any scripts before installing.