This is a surgical writing engine that takes a single viewpoint and dissects it layer by layer until you hit bottom. It enforces 1000-1500 word pieces in org-mode format, writes in conversational Chinese while avoiding AI tells and marketing speak, and saves to a timestamped Denote file. The core method is literal: cut once to expose what's underneath, cut again to go deeper, keep cutting until you reach bedrock or honest uncertainty. It's opinionated about everything from ASCII-only diagrams to banning phrases like "more deeply speaking" that announce depth instead of delivering it. If you write critical essays in Chinese and want a framework that forces you to think harder instead of cover more ground, this will either sharpen your process or irritate you into clarity.

skills.sh installs
6.7K
GitHub stars
7.7K
last updated
Aug 21, 2026

Install

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

Installs into .claude/skills of the current project.

Files

写作引擎

把一个观点讲清楚,让一个具体的人读懂它为什么成立、能用在哪里。用容易理解的案例说明关键关系;需要引入抽象概念时,先交代它能解释什么。

Workflow Routing

WorkflowTriggerFile
WriteEssay写文章、优化思想内容、展开或重写观点Workflows/WriteEssay.md

执行时完整读取工作流。分析和检查在内部完成,正文按内容需要展开,不逐项展示写作步骤。

成文契约

一篇文章围绕一个核心判断展开。涉及的其他机制、反馈和限制,都应帮助说明或修正这个判断:

  • 概念可指认:关键的抽象词都能用普通话解释,并对应到具体事物或现象。
  • 关系可运行:读者能说清谁在什么条件下作用于谁,产生什么结果。
  • 案例可映射:场景中的角色、条件和因果关系能对应原观点。
  • 判断可迁移:换掉人物和材料,关系仍成立;同时能指出一个失效边界。

例子帮助读者理解;换一个场景、检查一个反例,才能看出他是否掌握了这个判断。以上四项用于内部检查,不直接用作文章目录。

正文必须给出足够线索,让那个具体的人读完后能够:指出关键概念在现实中指什么,复述「条件—机制—结果」,并说出一个适用的新场景或一个失效边界。内部分析得再充分,这些线索也需要写进文章。

怎样解释一个观点

优先选择一个容易理解、关键因果关系完整的案例。从读者可能已有的解释出发,检查它会怎样判断、预期什么结果。若它解释不了某一步、预测有误,或忽略了某种代价,就把这个具体问题说明白。

随后引入能解决当前问题的概念,用普通话说明它增加了什么区分,再回到同一个案例,看解释或判断发生了什么变化。还有解释不清的地方,再继续补充。先说明概念的意思和用处,需要准确指称时再给出术语。

具体案例 -> 原有解释 -> 解释不清的地方
         -> 引入所需概念 -> 重新分析案例
         -> 说明完整关系 -> 检查其他场景与边界

这条路径用于帮助分析。正文可以先说判断,也可以从案例展开;原有解释已经够用时,直接说明原因,不必安排一次失败。解释结束时,读者应能看清各概念之间的关系,以及它们如何支持或修正最初的判断。

姿态与语言

写得清楚、准确,语气平实。可以有自己的判断和节奏,不必句句用力。

Sponsors

Details

Repository
github.com/lijigang/ljg-skills
First seen
Apr 16, 2026

More from lijigang/ljg-skills

All 25

Recommended

1juliusbrussee avatarcavemanjuliusbrussee/cavemanUltra-compressed communication mode. Cuts output tokens 65% (measured) by speaking like caveman while keeping full technical accuracy. Supports intensity levels618.3Kskills.sh installs
2mattpocock avatargrill-memattpocock/skillsA relentless interview to sharpen a plan or design.1.5Mskills.sh installs
3shadcn avatarimproveshadcn/improveSurvey any codebase as a senior advisor and produce prioritized, self-contained implementation plans for OTHER models/agents to execute. Strictly read-only on s40Kskills.sh installs
4obra avatarsystematic-debuggingobra/superpowersUse when encountering any bug, test failure, or unexpected behavior, before proposing fixes326.7Kskills.sh installs
5forrestchang avatarkarpathy-guidelinesforrestchang/andrej-karpathy-skillsBehavioral guidelines to reduce common LLM coding mistakes through explicit assumptions, simplicity, and verifiable success criteria.17.9Kskills.sh installs
6vercel-labs avatarfind-skillsvercel-labs/skillsHelps users discover and install agent skills when they ask questions like "how do I do X", "find a skill for X", "is there a skill that can...", or express int4.3Mskills.sh installs