Claude Skill MartBrowse skillsQuick linesLearn by videoTerminal guideWhat is a Skill?
 
← Back to list

Long-Form Web Novel Deconstruction (story-long-analyze)

A 7-stage pipeline that deconstructs a full-length web novel — golden first three chapters, characters, payoff rhythm, settings and prose style — and writes every artifact to disk.

Content & WritingIntermediate★ 6,326⑂ 932AI score 9/10Last updated: Aug 30, 2026

What it does

Give it a novel manuscript and it runs one deep-analysis pipeline (Stage 0–6):

  • Stage 0: synopsis + a chapter-boundary table (strips the table-of-contents block, disambiguates per-volume duplicate chapter numbers) that becomes the single source of truth for all later slicing
  • Stage 1: deep breakdown of the "golden three chapters", then auto-pauses, emits a quick-preview report and asks whether to continue
  • Stage 2: per-chapter summaries and plot points via 5–8 parallel chapter-extractor sub-agents, with grep-based hard checks, same-model retry and a sonnet upgrade retry
  • Stage 3–4: storyline/rhythm/emotion-module aggregation, worldbuilding, cheat-power and factions, character dossiers and relationship graphs
  • Stage 5–6: consolidated report plus a style profile (sentence length, punctuation, dialogue subtext, emotional alternation) with anchor excerpts

Everything lands under 拆文库/{book}/, and _progress.md lets you resume mid-run.

Who it's for

  • Web-novel and genre authors reverse-engineering what makes hits work
  • Editors and story consultants who need documented pacing and payoff maps
  • Users of the repo's story-long-scan → story-long-analyze → story-long-write pipeline

Examples

  1. Run /story-long-analyze, say "break down this book" with a path to the .txt — you get the golden-three-chapters preview first, then choose whether to go full depth
  2. Feed a 300-chapter serial and say "full deconstruction, don't ask" — it skips the checkpoint and runs Stage 2–6, resuming from _progress.md across sessions
  3. Re-run on a partially analyzed folder — existing character/setting files are used as a cross-validation baseline and only conflicts are surfaced for you to arbitrate

· · · Install guide · · ·

Try it now, no install

Paste this into Claude to use the skill without installing anything.

Read the instructions in this file and follow them to help me:
https://raw.githubusercontent.com/zenstory-ai/oh-story-claudecode/HEAD/skills/story-long-analyze/SKILL.md

What I want: (describe your task here)

If Claude can't open the link, open it yourself and paste the contents instead.

↓ If it works for you, download the ZIP below and install it. Then it runs on its own — no pasting each time.

Install in the Claude app (no terminal)
  1. Download the ZIP with the button below.
  2. In Claude, open Settings → Capabilities and turn on 'Code execution and file creation'. (one time)
  3. Go to Customize → Skills → + → 'Upload a skill' and upload the ZIP.
↓ Download ZIP
Install in Claude Code

Let Claude do it — paste this into Claude Code

Install the skill I found on Claude Skill Mart.
Copy the skills/story-long-analyze folder from the GitHub repo zenstory-ai/oh-story-claudecode into my ~/.claude/skills/story-long-analyze/.
When it's done, tell me in one line what this skill can do.

Install with a command instead

git clone https://github.com/zenstory-ai/oh-story-claudecode.git && mkdir -p ~/.claude/skills && cp -r oh-story-claudecode/skills/story-long-analyze ~/.claude/skills/

⚠ This is a third-party skill. Check the source repository before installing.

  1. Open a terminal in your working directory.
  2. Clone the repo: git clone https://github.com/zenstory-ai/oh-story-claudecode.git
  3. Copy the skill: mkdir -p ~/.claude/skills && cp -r oh-story-claudecode/skills/story-long-analyze ~/.claude/skills/
  4. Confirm the references/ folder (output-templates.md, material-decomposition.md, pipeline-ops.md, style-profile-*.md) came along — the pipeline loads its templates from there.
  5. For parallel extraction, run /story-setup as the repo instructs so the chapter-extractor agent is deployed; without it the skill falls back to serial mode automatically.
  6. Restart Claude Code, then run /story-long-analyze and supply the manuscript path (or paste the text).
  7. Find all outputs under 拆文库/{book title}/ in your project root.
View source on GitHub ↗License: MIT