論文倒讀法：給定一篇論文，遞迴找出它批判和改進的前序論文（最多5層），再找出它之後的最新進展，從源頭正向講述問題演化史。以問題為軸，費曼式講解每篇論文看到的問題和解法創新。
此技能名為「論文倒讀法」（paper river），旨在幫助開發者或研究者深入理解一篇論文的學術脈絡。使用者提供一篇論文後，技能會遞迴地往回追溯最多5層的前序研究（包括被批判或改進的論文），並同時找出該論文之後的最新進展。最終，它從問題的源頭開始，以正向時間順序講述問題的演化歷史，並採用費曼學習法（Feynman technique）來解釋每篇論文所識別的問題及其解決方案的創新點。此技能適用於Claude Code、Cursor、Codex等AI助手平台，特別適合當使用者想要理解論文之間的引用鏈、問題演進關係，或說出如「倒讀」、「論文溯源」、「論文脈絡」、「paper river」等關鍵詞時觸發。它通過分析論文間的批判與改進關係，提供了一個直觀、結構化的知識圖譜，讓使用者能快速掌握一個研究領域的發展軌跡。
npx skills add https://github.com/lijigang/ljg-skills/tree/main/skills/ljg-paper-riverInstalled? Explore more 研究與資料分析 skills: affaan-m/uspto-database, affaan-m/scholar-evaluation, affaan-m/literature-review · View all 6 →
論文倒讀法：給定一篇論文，遞迴找出它批判和改進的前序論文（最多5層），再找出它之後的最新進展，從源頭正向講述問題演化史。以問題為軸，費曼式講解每篇論文看到的問題和解法創新。當使用者分享一篇論文並希望了解其學術譜系、引用鏈、問題演化時，或者當使用者說出「倒讀」、「論文溯源」、「論文脈絡」、「paper river」、「paper connects」、「trace back」、「這篇論文的來龍去脈」、「論文演化」等關鍵詞時，即可觸發。同時，當使用者想要了解一個研究問題如何在多篇論文中演進時，也會觸發此技能。
This repo contains 19 individual skills — each has its own dedicated page.
Content caster (铸). Transforms content into PNG visuals. Seven molds: -l (default) long reading card, -i infograph, -m multi-card reading cards (1080x1440), -v editorial sketchnote (problem→failure→pivot→insight→naming, magazine + archive layout), -c comic (manga-style B&W), -w whiteboard (marker-style board layout), -b big-fonts attachment card (1080x1440, weathered 碑刻 style for 小红书). Output to ~/Downloads/. Use when user says '铸', 'cast', '做成图', '做成卡片', '做成信息图', '做成海报', '视觉笔记', 'sketchnote', '杂志', 'editorial', '漫画', 'comic', 'manga', '白板', 'whiteboard', '大字', '附件图', 'big fonts', '小红书卡片'. Replaces ljg-cards and ljg-infograph.
投资分析, 生成一份深度投资分析报告。不做传统投资分析——核心判断是项目是否是一台「秩序创造机器」。Use when user says '投资报告', '投资分析', '分析这个项目', '写投资报告', 'investment report', 'invest analysis', or provides entrepreneur conversation records wanting investment evaluation. Also trigger when user pastes or references meeting notes, pitch decks, or founder interviews and asks for analysis.
Deep concept anatomist that deconstructs any concept through 8 exploration dimensions (history, dialectics, phenomenology, linguistics, formalization, existentialism, aesthetics, meta-philosophy) and compresses insights into an epiphany. Use when user asks to explain, dissect, or deeply understand a concept, term, or idea. Triggers on '解剖概念', '概念解剖', 'explain concept', 'learn concept', '/ljg-learn'. Produces org-mode output.
Paper reader for non-academics. Reads a paper and tells it back as one continuous story — the life of the paper's core proposition (命题), told on a seven-beat spine (主角 / 困境 / 旧路 / 转折 / 解法 / 结局 / 内核): born in a bind on a base-rate ruler, crystallized as a bold conjecture, argued through mechanism and evidence, distilled into a new way of seeing, then walked out of the paper — life-tested and cashed into falsifiable predictions (检验). Output opens with a scannable 速读 card (一句话 / 大想法 / 只记三件事) that compresses the whole story three ways for the time-poor reader and the six-months-later self, then tells the full story. The job is storytelling that makes the paper land, not academic critique. Use when user shares an arxiv link, paper URL, PDF, or asks to analyze a research paper. Trigger words: '读论文', '讲论文', '把这篇讲给我听', '分析论文', 'paper', or when user shares an academic paper.
Paper workflow: read papers + cast 取景框 library cards in one go. Takes one or more arxiv links, paper URLs, PDFs, or paper names. For each paper, runs ljg-paper (generates org analysis) then ljg-library (distills the paper's 取景框 into a 2050 library card PNG). Use when user says '论文流', 'paper flow', '读论文并做卡片', '论文卡片', or provides multiple papers wanting both analysis and cards.
Cognitive atom: Plain (白). Rewrites any content so a smart 12-year-old groks it. Structure-free — form follows content. Use when user says '白话说', '说人话', '解释一下', 'plain', 'grok'.
演讲铸造器（Outline-Faithful）。基于 orgmode/markdown outline 层级 1:1 视觉化呈现——色块大字、ultra-bold 错位，原文不动只做美化。三档主题色 black/red/yellow（默认 black 或按 filetags 推断），可用 -r/-b/-y 显式覆盖；可用 --cyber 走黑底绿字 cyber-hacker 风。使用时用户会说：'讲这个'、'present'、'做成演讲'、'呈现一下'、'铸成演示'、'做个 slides'、'标语流'、'宣言体'、'slogan'、'manifesto'、'按 outline 美化'。输出单文件 HTML 到 ~/Downloads/。
把 ~/.claude/skills/ljg-* 里所有更新过的 skills 同步到 github repo (ljg-skills)，先推 master 分支（org-mode 输出风格），再切 md 分支（markdown 输出风格）做基础 markdown 化后推。Use when user says '/ljg-push', 'push skills', '推送 skills', '同步 skills', 'sync ljg', or whenever ljg-* skills get updated and need shipping. NOT FOR pushing non-ljg skills or arbitrary git repos.
信息提问机。给一篇文章/论文/书，把核心观点抽成 Q-A 对——Question 切要害，不教科书；Answer 简洁清晰，有形式化收口，逻辑链完整。读者顺 Q 链走过，每个 A 砸下一枚钉子，复现作者整套推理。Use when user says '问答', 'Q&A', 'QA', '提问', '抽取问题', '/ljg-qa', or shares an article/paper/book and asks for Q-A extraction. Triggers when the user wants ideas extracted not as a summary but as a sequence of incisive questions with answered. NOT FOR FAQ generation, glossary creation, or comprehension quizzes — this is intellectual scaffolding, not study aids.
给一个领域，找出背后真正撑着它的几根独立的力。十几个现象砍到不可再少的生成器——砍完能把现象一个个生回来，才算数。Use when user says '降秩', '找秩', '秩是什么', '这个领域靠什么撑着', '背后是什么', or wants to decompose any domain to its irreducible generators.
Reading companion agent. Accompanies user through any text (books, articles, essays, papers, news) with translation, structural annotation, deep questioning, and cross-domain insights. Detects language, translates English to Chinese (faithfulness-expressiveness-elegance), guides reader to understand the author and encounter real questions. Use when user says '伴读', '陪我读', '读这篇', 'read with me', 'companion read', or shares a text/URL wanting guided reading.
Relationship analyst combining structural diagnostics (5-layer framework) with psychoanalytic depth (transference, unconscious patterns, resistance). Guides users through dialogue to "see" the real structure of their relationship issues. Use when user says "关系分析", "分析关系", "relationship", "人际关系", or describes a specific relationship problem they want to understand.
Structured roundtable discussion framework with a truth-seeking moderator who invites representative figures for dialectical debate on any topic. Use when user says "圆桌讨论", "圆桌", "roundtable", "辩论", or wants to explore a topic through multi-perspective structured debate.
Skill map viewer. Scans all installed skills and renders a visual overview — name, version, description, category at a glance. Use when user says 'skills', '技能', '技能地图', 'skill map', '我有哪些技能', '看看技能', '列出技能', 'list skills'. Also trigger when user asks what skills are available or installed.
追本之箭——纵向深钻思维工具。给一个观点、现象或问题，像箭一样一路向下钻到不可再分的本质。Use when user says '想透', '追本', '本质是什么', '为什么会这样', '深挖', '钻到底', 'think deep', 'drill down', or wants to trace any idea/phenomenon vertically to its irreducible root. Also trigger when user provides a statement and wants depth analysis, not breadth survey.
Deep travel research workflow for museums and ancient architecture. Input a city name, auto-generates structured knowledge document (org-mode) + portable reference cards (PNG). Covers historical background, museum highlights, archaeological significance, and architectural heritage. Use when user says '旅行研究', '博物馆功课', '古建功课', 'travel research', '出发前功课', or provides a city name with intent to do deep cultural travel preparation.
Deep-dive English word mastery tool. Deconstructs a single English word into core semantics and epiphany. Use when user asks to explain/master a specific English word.
Word flow: deep-dive word analysis + infograph card in one go. Takes one or more English words, runs ljg-word (generates deep semantics analysis) then ljg-card -i (generates infograph PNG). Use when user says '词卡', 'word card', 'word flow', or provides English words wanting both analysis and visual card.
写作引擎。像手术刀剖开一个观点，一层层剥到底。1000-1500 字。
USPTO patent and trademark data workflow for official record lookup, PatentSearch queries, TSDR checks, assignment data, and reproducible IP research logs.
Structured scholarly-work evaluation for papers, proposals, literature reviews, methods sections, evidence quality, citation support, and research-writing feedback.
Systematic literature-review workflow for academic, biomedical, technical, and scientific topics, including search planning, source screening, synthesis, citation checks, and evidence logging.
The agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode, Cursor and beyond.
Agent skill repository discovered by 10x-chat research.
Review backend code for quality, security, maintainability, and best practices based on established checklist rules. Use when the user requests a review, analysis, or improvement of backend files (e.g., `.py`) under the `api/` directory. Do NOT use for frontend files (e.g., `.tsx`, `.ts`, `.js`). Supports pending-change review, code snippets review, and file-focused review.