ljg-read

ljg-read is a skill for Claude Code from lijigang/ljg-skills. It costs 106 tokens per session (3,187 once invoked), scanned A, original, MIT.

A reading companion for books, articles, essays, papers, and news. It can translate English into Chinese, explain a text’s structure, annotate important ideas, and ask questions that deepen understanding.

In plain words
What is it for?
Use it to translate and study English writing, map its structure, examine difficult concepts, and develop deeper questions across different fields.
Why use it?
It helps readers get past language barriers and see the author’s argument, evidence, transitions, and hidden assumptions. Its aim is guided understanding rather than a simple summary.

Skill for Claude Code

Written for Claude Code: user-invocable in frontmatter.

Part of the ljg-skills plugin — 22 skills shipped together

Good fit Use it to translate and study English writing, map its structure, examine difficult concepts, and develop deeper questions across different fields.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/lijigang/ljg-skills/ljg-read
About the project

ljg-skills is a collection of custom Codex skills for tasks such as learning, writing, reading, relationship analysis, image creation, and investment analysis. Codex users install selected skills or the whole collection through a skills command-line interface. The catalogue entries are the collection's individual skills, plugin, and instruction.

lijigang/ljg-skills · 7,340 stars · on GitHub

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Any agent
npx skills add lijigang/ljg-skills --skill ljg-read
Clone the repo
git clone --depth 1 https://github.com/lijigang/ljg-skills

Made for: Claude Code.

Or install ljg-skills, the plugin that ships this one along with the rest of its 22 skills.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for ljg-read

README.md
[![agentmods](https://agentmods.dev/badge/skills/lijigang/ljg-skills/ljg-read/github.svg)](https://agentmods.dev/skills/lijigang/ljg-skills/ljg-read)
Your own site
<a href="https://agentmods.dev/skills/lijigang/ljg-skills/ljg-read"><img src="https://agentmods.dev/badge/skills/lijigang/ljg-skills/ljg-read/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for ljg-read

Your own site · 80×15
<a href="https://agentmods.dev/skills/lijigang/ljg-skills/ljg-read"><img src="https://agentmods.dev/badge/skills/lijigang/ljg-skills/ljg-read.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 106 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 3,187 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe. Third-party audits
  • Socket pass 13 Apr 2026
  • Snyk warn 13 Apr 2026
  • NVIDIA SkillSpector pass 7 Sept 2026
How audits are shown
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00106 $0.03187
Opus 5 $0.00053 $0.01594
Sonnet 5 $0.00021 $0.00637
Haiku 4.5 $0.00011 $0.00319

Measured 12d ago against content hash 07c9138e6675, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-11, from the pricing page.

Security

Grade A, and why

ljg-read scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 12d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

skills/ljg-read/SKILL.md · 274 lines

How it starts

The opening of the file, as written. The whole thing — 274 lines — stays where its author put it; the contents beside it link to each section on GitHub.

ljg-read: 伴读

不是替你读,是陪你走进去。扫清语言障碍只是起手式,真正的活儿是让你撞见自己从没想过的问题。

核心理念

  • 翻译是再生产,不是搬运——信是不歪,达是懂了,雅是住下来了
  • 伴读是脚手架,最终要拆——读者被激活才算有效
  • 最好的伴读不回答问题,而是制造那个让你皱眉的问题

格式约束

Org-mode 语法

  • 加粗用 *bold*(单星号),禁止 **bold**
  • 标题层级从 * 开始,不跳级

ASCII Art

所有图表用纯 ASCII 字符。禁止 Unicode 绘图符号。

语言

默认中文输出。英文原文在翻译和碰撞环节以中英并列形式保留。

执行流程

0. 接收文本

  • URL -> WebFetch 或 markdown-proxy 获取内容
  • PDF -> Read(注意 pages 参数限制)
  • 本地文件 -> Read
  • 用户粘贴文本 -> 直接使用

获取后,检测语言。中文文本跳过翻译步骤,直接进入结构分析和碰撞。英文文本进入完整流程。

1. Phase 0: 全局地图(一次性,Agent 独立完成)

通读全文,产出三样东西:

(1) 一句话摘要 — 这篇文章在说什么,一句话锚定。

(2) 段落分类 — 将全文段落标记为三类:

  • [骨] 骨架段:承载核心论证/核心观点
  • [肌] 肌肉段:展开论证的证据、例子、数据
  • [筋] 筋膜段:过渡、连接

标注为"Agent 判断",读者可覆盖。

(3) 全文结构地图 — 论证单元之间的关系,用简短的 ASCII 图或缩进列表呈现。

(4) 五维段前扫描(内部决策,不输出给读者)— 对全文做初步判断:

  • 语言密度(术语多寡、句法复杂度)
  • 文本性质(论证/叙事/抒情/说明,逐段可能不同)
  • 文化距离(中英文化错位程度)
  • 论证张力(可追问的前提/跳跃)
  • 类比潜力(跨领域同构可能性)

向读者呈现:一句话摘要 + 结构地图 + 段落分类概览。

2. Phase 1: 逐段翻译推进

节奏因段落分类而异:

[骨] 骨架段 — 精读

翻译三层递进输出:

直译层(信):严格对应原文,逐句翻译,不加不减。关键术语中英并列。

意译层(达):用自然中文重述整段意思。调整语序,补充隐含逻辑,拆长句。不做翻译腔。

点睛层(雅,按需):仅在以下三种情况触发——

  • 中英文概念有系统性错位(如 freedom vs liberty,中文"自由"盖不住)
  • 作者用了领域内行话,圈外人无法自查
  • 句子有双关或文化典故,字面翻译丢信息

翻译后自动停下,进入 Phase 2。

翻译操作细节
  • 语义切段:不按原文自然段切。按语义单元重新切段——一个论点一段,一个论据一段。通常原文一个自然段切成两到三个语义段。硬约束:不切断完整论证。
  • 术语策略:首次出现给「英文原词(中文译法)+ 一句话定义」;后续出现只给中文译法加括号英文。
  • 文化翻译:遇到中英写作传统差异时主动点破——
    • 英文让步结构("However, one might argue...")不等于作者立场动摇
    • 英文散文的 understatement 传统:语气越平淡,作者越认真
    • 英文新闻的倒金字塔 vs 中文的起承转合
[肌] 肌肉段 — 流读

翻译连续呈现,不自动停下。末尾标注:"以上 N 段在支持骨架段 X 的论证"。

[筋] 筋膜段 — 跳读

一句话带过:"作者从 A 过渡到 B"。

结构标注

每段翻译后附一句话标注:此段在全文论证中的角色——"核心论点""对第二段的反例""让步之后的转折""证据铺垫"等。

3. Phase 2: 骨架段深入

3a. 注疏(先问后给)

先判断读者对当前概念的状态(模糊 / 已接受未检验 / 已理解不知重要性),然后:

先问:"这个概念让你想到了什么?"或更具体的指向性问题。

  • 读者能自己连接 -> 确认/微调,进入碰撞提问
  • 读者连不上 -> 给一条注疏,三种光选一:
    • 同构(侧光):别的传统里谁说了同样的事
    • 对手(逆光):最强的反对是什么
    • 源流(背光):从哪来,改变了什么

约束:

  • 每次只给一条,选杀伤力最大的
  • 引用外部文本时中英并列
  • 读者可说"还有吗"追加,最多三条后收束
3b. 碰撞提问

核心问题(诊断+催化二合一):"作者这段最想说服你接受的一个点是什么?你接受吗?"

Read the full file on GitHub · 274 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 12d ago First seen · 274 lines · 106 tokens per session scan A 07c9138e6675

Subscribe to this mod's changes

ljg-read is a skill published in the GitHub repository lijigang/ljg-skills (7,340 stars, last pushed 3d ago), licensed MIT. It adds 106 tokens to every session and 3,187 once invoked, about $0.0005 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Related

Other skills, from other repositories

toolbox-talk-generator

Generate safety toolbox talks for construction crews. Create contextual safety briefings based on weather, work activities, and recent incidents. Support multiple languages.

datadrivenconstruction/DDC_Skills_for_AI_Agents_in_Construction · 33 tokens

vietnamese-education-copy

Writes and reviews native-quality Vietnamese (vi-VN) educational, school, academic, and EdTech communication — K-12 report-card remarks and học bạ entries, sổ liên lạc entries and school-to-parent broadcasts, pedagogical nudges, diagnostic pronunciation feedback, adaptive assessment, parental analytics reports…

trussary/vietnamese-language-skill · 217 tokens

lov-subtitle-freedom

Create learner-friendly English subtitles with level-aware glosses and optional spoiler-safe subtitle sidecars.

lovstudio/skills · 25 tokens

apple-design

Apple's approach to interface design and fluid, physical motion, translated for the web. Use when building or reviewing gesture-driven UI, spring animations, drag/swipe/sheet interactions, momentum and interruptible transitions, translucent materials and depth, typography (optical sizing, tracking, leading)…

classroomio/classroomio · 80 tokens

stage-and-commit

Stage changed files and create a commit following the repo's translation, formatting, and git workflow rules. Use when the user asks to stage, commit, or both.

classroomio/classroomio · 38 tokens

01-k12-languages

A comprehensive language learning tutor for K-12 students covering English, Chinese, and other world languages. This skill transforms the AI agent into an immersive language partner that develops all four core competencies — reading, writing, listening, and speaking — through communicative practice grounded in…

gabrielmoreira/agent-skills-mirror · 0 tokens