Ruxi Skill

入戏.skill —— 把一本书变成你能玩进去的视觉小说（galgame）。给它一本书（EPUB/TXT，或一个书名），它产出一个双击即玩的单文件 HTML 游戏：你扮演原著里被略过的边角人物，在原文真实存在的抉择路口上重新选一次。每个分支点都标注原文出处；不做好感度，做「理解度」——通关告诉你这本书你读懂了多少；结局里永远有一条是书里真实发生的事。零素材依赖（纯 CSS/SVG 立绘），版权判定写进流程。触发词：「把 X 做成游戏」「X 改编成 galgame / 视觉小说 / 文字游戏」「书改游戏」「入戏 X」「我想玩进 X 这本书」「novel to visual novel」「book to galgame」「make X playable」。

1 stars 1 forksWriting & Content

Install

skills CLI (Claude Code, Cursor, Codex and more)

npx skills add swaylq/ruxi-skill --skill ruxi-skill

Manual install

git clone --depth 1 https://github.com/swaylq/ruxi-skill.git /tmp/ruxi-skill
mkdir -p ~/.claude/skills
cp -r /tmp/ruxi-skill/. ~/.claude/skills/ruxi-skill

Personal skills live in ~/.claude/skills. Use .claude/skills in a project folder to share a skill with your team.

From the SKILL.md

# 入戏 · 把一本书变成你能玩进去的一场戏

> 你不是读者。你是书里那个只有三行字的人，
> 站在原著真实发生过的抉择路口上，替她重新选一次。

## 这个 skill 干什么

输入一本书，输出一个 `index.html`——双击即玩、离线可玩、可发给朋友的视觉小说。
它和「让 AI 写个 galgame」的区别是五条铁律（详见 `references/adaptation.md`）：

1. 不覆盖全书，只取五到八场戏
2. 玩家默认演原著里被略过的边角人物
3. 每个分支点必须指得回原文的一句话——找不到出处的分支砍掉
4. 不做好感度，做理解度：选项是解读，通关结算「你读懂了多少」
5. 结局里必须有一条是原著真实发生的事，标为正典

完整样例：`demos/metamorphosis/`（《变形记》——你是萨姆沙家十六岁的帮厨女孩）。

## 工作流：五阶段

### 阶段 0 · 版权判定（先于一切）

问用户或查证两件事：**作者卒年、首次发表年**。按 `references/adaptation.md` §6 判档，
写进 `meta.book.pd_status`。三档：全球公版 / 美国公版 / 有版权（仅限本地自娱，
构建器自动注水印）。**译本有独立版权**——引用一律引原语言初版，译文自译。
拿不准按有版权处理，并明确告知用户这不是法律意见。

### 阶段 1 · 拆书选戏

```bash
python3 tools/scan_scenes.py book.epub            # 英文书
python3 tools/scan_scenes.py book.txt --names "甲,乙,丙"   # 中文书必须给人名
python3 tools/scan_scenes.py book.epub --json     # 给自己读的全量数据
```

压强表出来后，精读入选章 + 人物首次登场章。压强公式会漏「沉默的重场」和
「信息爆点」，按方法论 §1 人工补。选定 5-8 场，排成一条线。

### 阶段 2 · 定角

按方法论 §2 找边角人物（grep 仆人/信使/邻居/无名者），验三条标准，
**把「玩家演谁」和「这个角色自带的中心问题」摆给用户确认**，再动笔。

### 阶段 3 · 写本

- 剧本 = 一个 JSON，结构定义在 `references/script-schema.md`，逐字段照填，不得加字段
- 写前先做文风档案（方法论 §5），旁白台词全部按档案写
- 每个 option 的 source 从原文精确摘句（≤60 字）；发明的场景诚实标注
- 理解度计分自查三问（方法论 §4）

### 阶段 4 · 构建与自测

```bash
python3 tools/build_html.py script.json out/index.html   # 校验 + 构建
python3 tools/build_html.py script.json --check          # 只校验
```

校验器抓：goto 断链、cast 引用错误、canon 结局不唯一、option 缺出处、超长台词。
构建后**必须自己玩一遍全部路线**（浏览器打开，或无头跑一遍分支图），
确认每条 goto 可达、结算页判词正确，再交付。

### 阶段 5 · 交付

Read the full skill on GitHub

Details

Skill name: ruxi-skill
Repository: swaylq/ruxi-skill
License: MIT
Repo updated: 1 September 2026
Listed in: karanb192/awesome-claude-skills

FAQ

What does the Ruxi Skill skill do?

How do I install the Ruxi Skill skill?

skills CLI (Claude Code, Cursor, Codex and more): npx skills add swaylq/ruxi-skill --skill ruxi-skill. Skills go in ~/.claude/skills (personal) or .claude/skills (project), and Claude loads them automatically when a task matches the description.

Is Ruxi Skill popular?

It lives in swaylq/ruxi-skill, which has 1 GitHub stars and 1 forks, and is included in 1 curated skill list.

Related skills

Apify Linkedin Post Engagement

johnisanerd/claude-skill-linkedin-post-engagement

Analyze LinkedIn post engagement with the Apify LinkedIn Posts API Actor (johnvc/linkedin-posts-api). Collect a profile's posts or a list of post URLs, then measure reactions, comments, and shares alongside author reach (followers, headline) to see what performs. Use when the user wants to analyze LinkedIn post engagement, measure engagement rate, find a creator's or a competitor's top-performing posts, benchmark engagement across profiles, track post engagement over time, or build a LinkedIn engagement report. Returns one row per post with numLikes, numComments, numShares, datePosted, and a r

Writing & ContentIn 1 list

Diasporic Intelligence

MinistaJazz/diasporic-intelligence

Use when writing, reviewing, citing, or implementing Diasporic Intelligence, a source-credited framework also described as Afrotemporalism in Action. Preserves source credit, consent governance, provenance, revocation, lineage boundaries, and the distinction from generic cultural AI, biometric extraction, and ancestor impersonation.

Writing & ContentIn 1 list

Goethecoach

janosszaboaipm-design/goethecoach-claude-skill

German exam coach for the Goethe-Zertifikat (A1–C2) by GoetheCoach. Generates exam-style Lesen, Schreiben and Sprechen practice tasks, evaluates learner answers against the official criteria (Aufgabenerfüllung, Kohärenz, Wortschatz, Strukturen), explains errors through the learner's native language, runs follow-up tutoring and builds study plans up to the exam date. Answers questions about the exam (format, scoring, Leitpunkte, dates, digital exam, Goethe vs. telc, failing) and everyday German for life in Germany from the bundled GoetheCoach article library, with links for further reading. Use

Writing & ContentIn 1 list

Humanizing Writing

AshwinSathian/humanize-writing-skill

Use when drafting prose of a paragraph or more for a human reader (docs, READMEs, reports, emails, blog posts, cover letters, release notes, PR descriptions, commit bodies, long explanations in chat), or when asked to humanize text or make it sound less AI-generated, robotic, or like slop. Not for proofreading a person's own writing or for evading AI detectors.

Writing & ContentIn 1 list

Publora Post Ideas

publora-team/publora-post-ideas

Suggests three different angles for a social post and writes the one you pick, instead of guessing at a single draft. Use when the request is open-ended — "write me a post", "I need content ideas", "I don't know what to post today", "give me something for LinkedIn" — or when a draft was rejected and a different approach is needed. Pairs with publora-publishing, which does the actual scheduling. Works on its own; with Publora connected it also picks angles against the user's real feed and schedules the result.

Writing & ContentIn 1 list

Okf

mattjoyce/okf-skill

Create, author, validate, and navigate Open Knowledge Format (OKF) bundles — vendor-neutral knowledge represented as a directory of markdown files with YAML frontmatter. Use when asked to create an OKF bundle, write a concept document, build a knowledge bundle/catalog as markdown, produce index.md or log.md for a bundle, check OKF conformance, convert metadata or docs into OKF, or work with anything described as OKF, a "knowledge bundle", or "metadata as markdown".

Writing & ContentIn 1 list

Browse the most popular Claude skills, all writing & content skills, or write your own with the skills generator.