Story Import Skill (Oh Story): Turn an Existing Novel into a Writing Project

Rebuilds an existing Chinese web novel, finished or half written, into a project folder with settings, outlines and tracking that Claude can continue writing.

Installs: 16.4K; on skills.sh
Stars: 8.2K; on GitHub
Files: 17; besides SKILL.md
Updated: Nov 2026; last repo commit

Story Import makes Claude read a novel you have already written, finished or half done, and rebuild it as a writing project with settings, outlines, chapters and tracking files. After that you can keep writing with the other Oh Story skills, for example by asking for the next chapter. It works in Chinese: SKILL.md and its triggers (such as 导入小说) are written in Chinese.

Claude first confirms the title, genre, platform, chapter range and whether the last chapter is complete. It then runs the long or short novel analysis skills, moves the results into the project folders and reports what you should check. Long novels also get outlines worked back from each chapter and a tracking folder built by a Python script.

Good for

Web novel authors who want Claude to continue a book they already started
Writers moving a manuscript into the Oh Story workflow
Authors who read and write Chinese, since the skill works in Chinese

Install story-import

Claude Code

Copy this prompt.

Install the story-import skill in Claude Code. Run this command:

npx skills add zenstory-ai/oh-story-claudecode -y -g

Then check that it loaded and tell me how to use it.

2
Paste it into Claude Code. It shows the change and asks before running it.

Requires Node.js.

story-import calls story-long-analyze and story-short-analyze, and the README says to run /story-setup in your writing project after installing. Install the whole Oh Story set, not story-import alone. Everything in it is written in Chinese. Drop -g to install into the current folder only. Re-run the same command to update.

From the project README. Source.

What's in the skill

When it loads

逆向导入已有小说。将已写好的小说（半成品或完本）反向解析为标准项目目录结构，兼容 story-long-write / story-short-write 后续写作流程；内部复用 story-long-analyze / story-short-analyze 的拆解管道，按篇幅自动分流。触发方式：/story-import、「导入小说」「反向解析」「导入」「把我的书导进来」。

Sections in SKILL.md

名词与目录边界（全流程硬约束）
时刻表与交接
Phase 1：确认来源与范围
Step 1：确认意图
Step 2：拿到原文
Step 3：基本信息确认
Step 4：写作环境检测
Step 5：写导入记录
进度
Phase 2：深度分析
Phase 3：结构迁移
Phase 4：追踪初始化（仅长篇）
Phase 5：汇报与激活
流程衔接

Key facts

Author: zenstory-ai (Oh Story)
License: MIT
Installs on skills.sh: 16,417 (Nov 2, 2026)
GitHub stars: 8,223 for the whole zenstory-ai/oh-story-claudecode repo (Nov 2, 2026)
Last commit: Nov 1, 2026
Skill version: 1.0.0 (SKILL.md front matter)
What is inside: 18 files including SKILL.md: 12 reference guides and 5 scripts (4 Python, 1 Node.js)
Needs: story-long-analyze or story-short-analyze from the same repo; the README says to run /story-setup first
Language: Chinese (SKILL.md, triggers and reports)
Works in: Claude Code, Codex CLI, Google Antigravity, OpenCode, ZCode, OpenClaw, Reasonix (README, for the Oh Story set)

What it includes

Asks for the text, title, genre, platform, length and last-chapter status
Claude takes a file path or pasted text, detects chapters and word count, and asks you to confirm. It saves the result in 导入记录.md so a new chat can resume.
Try: “导入小说 ~/novels/我的书.txt”
Write · Phase 1: Confirm source and scope
Runs the long or short analysis skill and saves to 拆文库/
Chosen by length. Needs story-long-analyze or story-short-analyze. Results go to 拆文库/{book name}/ and are kept as the data source.
Write · Phase 2: Deep analysis
Moves the analysis into settings, outline and chapter folders
Long novels become 设定, 大纲 and 正文 folders. Short novels become three single files. If the original has no volumes, you decide how to split them.
Write · Phase 3: Structure migration
Works back a detailed outline for each chapter, batch by batch
Uses scripts/build_outline_brief.py to build a task pack for each batch. scripts/check-outline-contract.js (Node.js) checks the outline structure.
Runs code · Outline reverse (long novels)
Builds the 追踪/ folder with tracking_commit.py init, then check
A Python script writes the tracking state and its Markdown views for plot threads and characters. It also removes character-state files it no longer expects.
Delete · Phase 4: Tracking setup (long novels)
Self-checks the project, reports what to review, sets .active-book
The report lists up to five things for you to confirm and tells you how to continue, for example with /story-long-write and the next chapter number.
Write · Phase 5: Report and activate
Points to story-long-write, story-short-write, story-review and story-setup
These are separate skills in the same Oh Story repo. Import does not write new chapters itself.
Read · Hand-offs

From the official docs

Permissions

Medium risk

Story Import copies your manuscript into several project folders and runs Python and Node.js scripts that write and delete files in the book folder. Work on a copy of your manuscript and keep the original file safe.

What it can do

Read your data
Create and change things
Delete things
Read-only mode

Watch out for

The setup step installs hooks and agents in your project
The README says /story-setup deploys hooks, agents and references for your coding agent, and those hooks run checks automatically. Read what it adds before you accept it. github.com
Scripts write and delete files in the book folder
tracking_commit.py writes the tracking files and deletes character-state files it does not expect. Keep the project under version control. github.com
Your whole manuscript is copied around
SKILL.md saves pasted text to .story/work and backs up the original into 拆文库. Do not import text you cannot store in your project folder. github.com

Technical details

Scopes	Reads your manuscript file or pasted text Writes 拆文库, 设定, 大纲, 正文, 追踪 and .story/work folders Runs Python and Node.js scripts from the skill folder

Questions

What does the story-import skill do?

It reverse-engineers an existing novel, finished or half written, into a project folder with settings, outlines, chapter text and tracking, so you can continue with the other Oh Story writing skills.

How do I install story-import?

Use the install box at the top of this page. It installs the whole Oh Story set, because story-import reuses the analysis skills. Then run /story-setup in your writing project and start a new session.

Does it work in English?

It is built for Chinese web novels. SKILL.md, the triggers and the reports are written in Chinese, and it names platforms such as Qidian, Fanqie and Jinjiang. The README gives an English version of the project page, but the skill text itself is Chinese.

Does it need an API key?

The README says Oh Story runs inside the coding agent you already use and uses that agent's model. The one exception it names is story-cover, which calls GPT-Image-2. Story-import is not that skill.

Is it safe to run?

It reads your manuscript and writes many files and folders in your project, and the setup step adds hooks. Work on a copy and use version control.

What is the difference from story-long-analyze?

story-long-analyze takes a book apart into an analysis. SKILL.md says that if you only want the analysis, use the analyze skill and stop at 拆文库. Story-import goes further and rebuilds a project you can keep writing in.

What do I do after importing?

The skill says to continue with /story-long-write and the next chapter number, or /story-short-write to finish a short story idea. Run /story-review if you want a quality check.

Information

updated Nov 2, 2026

Repository: zenstory-ai/oh-story-claudecode
Folder: skills/story-import

zenstory-ai/oh-story-claudecode skills.sh

Sources

Docs, issue trackers and discussions this page is built from.

Show all 4

github.com · github.com
github.com · github.com
github.com · github.com
github.com · github.com