How authors write novels in Claude Code with custom folders, CLAUDE.md and tracking files—and how Maliang Writer compares with built-in story archives.
Claude Code can write novels, and it is very capable: it reads and writes a whole book’s folder, checks continuity across files and makes batch edits. In public practice, authors design the folders, write CLAUDE.md, maintain a setting ledger and work in a terminal or editor plugin, which suits authors who are comfortable on the command line. Maliang builds the table of contents, character cards, story archives and revision comparison into a web editor, and the author starts writing from the chat.
An amateur author who works in back-end development wants to write a 500,000-character science-fiction novel with Claude Code. Following an open-source project online, he creates the folders chapters, characters, outline and tracking, writes the style, forms of address and a rule of “update the tracking file after every chapter” in CLAUDE.md, and keeps versions in Git. The workflow suits him well, and he knows exactly what every file is for. If he wanted to recommend the same workflow to friends who don’t write code, he would first have to teach them to set up the environment, read the folders and read diffs. What Maliang does is build this structure directly into a web editor.
On narrow screens, scroll sideways to see all columns.
| Dimension | Maliang Writer | Claude Code | Notes |
|---|---|---|---|
| Product form | ✅ Web editor plus AI chat, ready to use in the browser | ✅ Terminal CLI, plus VS Code / JetBrains plugins, a desktop app and a web version | |
| Manuscript structure | ✅ Volumes, chapters and chapter outlines are in the table of contents; characters, settings, foreshadowing and storylines are cards; the master outline is stored separately; style samples are documents | ⚠️ No official material found on novel-specific management; in public practice authors build their own chapters, characters, outline and tracking folders | |
| Where writing rules go | ✅ Material such as the master outline and style samples is read by AI as needed when writing | ✅ CLAUDE.md and Auto memory provide cross-session memory, and the author writes the rules | |
| Tracking setting changes | ✅ After a chapter is written, story archives are organized: character states, relationships, locations, foreshadowing; anything that doesn’t match is listed as pending confirmation; can be turned off | ⚠️ In public practice authors mostly design tracking files or ledgers and have AI update them by rule | |
| Reading and writing the whole project | ✅ AI can read and write this book’s chapters, characters, settings and documents, and changes go straight back into the editor | ✅ The official site says it “understands your entire codebase” and can run tasks across files | |
| Change comparison | ✅ Red deletions and green additions on the original in the editor, each of which can be kept or restored | ✅ The VS Code plugin and the desktop app can show diffs | |
| Undo and redo | ✅ Undo the current round, branch from before the round, or edit and resend | ✅ A checkpoint is created automatically with every prompt, the latest 100 are kept, and you roll back with /rewind | |
| Old settings once you reach a new volume | ✅ Characters, places and foreshadowing that only appeared in earlier volumes are moved into the archive; AI no longer consults them when writing new chapters, but they can still be looked up when needed | ⚠️ No official material found (the author can achieve it by designing their own folders) | |
| Genre reference | ✅ When starting a book, consult the character and structure breakdowns of more than 1,500 popular serials in the public book library; chapter endings are benchmarked against about 56,000 popular chapter endings | ⚠️ No official material found | |
| Price | ✅ Free sign-up to try; credits are deducted by usage | ✅ Not included in the Free tier; included from Pro at $20/month (checked 2026-10) | |
| Regions | ✅ A Chinese-language web product for Chinese-language authors | ❌ The official supported-region list does not include Mainland China, Hong Kong or Macau, and the restriction covers Claude Code |
✅ Confirmed in official material or in the product ⚠️ Partly supported, or not stated in official material ❌ Officially not supported · Competitor information comes from their public material.
Chapters, chapter outlines, characters, settings, foreshadowing and storylines each have a place in the editor, and what the author sees is a table of contents and cards, not a file tree. The master outline is stored separately, style samples are documents, and AI reads them as needed when writing.
After a chapter is written and this stretch of writing ends, the background organizes this chapter’s archive: character states, relationships, locations, foreshadowing, and lists anything that doesn’t match as pending confirmation. You don’t have to write a rule for “update the ledger after every chapter”. It can be turned off.
When AI edits the body text or a document, changes are marked on the original as red deletions and green additions, and each can be kept or restored. If you are unhappy with a whole round, you can undo the current round, branch (sharing the same manuscript) or edit and resend.
Claude Code reads and writes a whole folder and runs tasks across files, with a 1M context, checkpoint rollback and diff viewing. Some authors have used it for continuity checks and batch edits across 113 files, and there are open-source projects that provide a complete novel folder layout and rules for it, compatible with Claude Code, Codex CLI and other tools. For authors comfortable on the command line it offers great freedom: folders, rules and check scripts can all be designed as you see fit.
Every product suits some scenarios; the advice here covers only one: serialized long-form fiction in Chinese.
For this page, record in particular how much work the author has to do to prepare folders, writing rules, material updates and revision checks. Don't compare only the prose of a single generation, and don't assume in advance that folder-based tools must be harder to use.
Below is a same-task trial checklist, not a finished comparison experiment. You can use the original teaching material “Jiehuo City” or a manuscript you have the right to handle; record the preparation, the output and the author’s edits for each tool.
Give both tools the same protagonist, core rules and reading preferences; compare the opening directions first, then organize the master outline, the first volume outline and the outlines for the first three chapters.
Check: Can it tell the roles of the three outline levels apart? Can the author confirm a plan before it continues? Were settings that hadn't been settled yet quietly written in as fact?
Use the same book-opening kitUse the same old manuscript you have the right to handle, specify how the last chapter ends and the goal of the next chapter; try writing one scene first, without changing the previous twenty chapters.
Check: What material did you prepare? Did character positions, items and open threads carry over? How are candidate drafts told apart from adopted ones?
See the import-and-continue task checklistPick a real rule that affects earlier chapters; first ask it to list the affected chapters and the evidence, then let the author choose the scope of the change.
Check: Does it say which parts it did not check? Can you review the changes first? Did it wrongly edit text outside the task or write an unadopted plan into the material?
See how to change a settingActually try keeping the text, the order of volumes and chapters, character profiles and world rules, and check whether the output is suitable for further editing; list what cannot be taken along.
Check: Are the format and order recognizable? What material is inside the platform account, what is in local files, and what is in the export? Do not infer usability from a marketing page alone.
Copy the task and result record cardRecord the models, plans or account conditions you actually used, and the time spent preparing materials and editing the text. Leave blank anything you have not tried or cannot confirm; do not let a feature name or a single generation stand in for the full experience.
Yes. Claude Code reads and writes a whole folder, and some authors use it to write long novels, check continuity and make batch edits. It is a tool designed for programming, so for writing novels the author has to build the folder structure, rule files and setting ledger.
A common approach in public practice: keep settings, characters, outlines (master outline, volume outlines, chapter outlines) and body text in four separate places, add a status or tracking file as the overall source of truth, put hard rules in CLAUDE.md or AGENTS.md at the root, and use Git for versions.
Public projects usually include: style and point-of-view requirements, forms of address for characters, which files are the source of truth, which tracking files to update after each chapter, and settings that must not be changed. The more rules there are, the more carefully they must be maintained.
Some authors in public practice mention style drift and setting drift, and say you need to feed in the relevant settings, the previous few chapters and sample passages for each chapter, and maintain tracking files. In Maliang, the background organizes story archives after a chapter is written and lists anything that doesn’t match as pending confirmation.
No. Claude’s free tier does not include Claude Code; it is included from Pro ($20/month) and above (checked against the official pricing page in 2026-10).
Claude Code also has a desktop app and a web version, but most public practice for writing novels is based on local folders and rule files. One author describes it as “more powerful (and more intimidating)”. If you don’t want to deal with file structure, you can use a web writing tool like Maliang directly.
The Claude web version uses Projects to hold material and writes in conversation; Claude Code reads and writes all the files in a local folder and can make batch edits across files, but you have to organize the files yourself. For a comparison of writing novels in the Claude web version, see the “Claude novel writing vs Maliang Writer” page.
Anthropic’s official supported-region list does not include Mainland China, Hong Kong or Macau, and states that the restriction covers Claude Code.
Maliang’s AI can read and write this book’s chapters, characters, settings and documents, with changes going straight back into the editor and body-text changes reviewable one by one. When a change touches many chapters, the background archive organizing will ask for your confirmation first.
Claude suits discussing plot, drafting passages and revising, and Projects keeps settings and instructions together. In a long serial, which earlier text goes with each chapter, where to edit when a setting changes and what this version changed are all for the author to arrange. Maliang turns characters, settings and foreshadowing into cards, organizes story archives after each chapter is written, and shows AI changes right in the editor, where you can keep or restore each one.
Read the comparison →Cursor can write novels: the editor lets you see files and live diffs, the Rules system is detailed, and the choice of models is wide. In public practice, authors build their own chapter folders, write rules and maintain settings and a ledger, and once there are many chapters they also have to deal with file ordering and merging. Maliang builds the volume-and-chapter table of contents, character cards, story archives and revision comparison into a web editor.
Read the comparison →Codex CLI is open source and runs in a local terminal; it reads and writes a whole project and runs scripts, with system-level isolation and approvals. Writing a novel with it, the author builds the folders, writes AGENTS.md, maintains a ledger and uses Git for versions, and views changes as a text diff in the terminal. Maliang builds these into a web editor: a volume-and-chapter table of contents, character cards, story archives and revision comparison you review one by one.
Read the comparison →Coding agents are very capable: they read and write across files, make batch edits and run scripts, and technical authors can build very flexible writing workflows. The price is that the folders, rule files, setting ledger and version management all have to be designed and maintained yourself, and the interface is aimed at programmers. Maliang turns these into the cards, table of contents and editor that authors are used to.
Read the comparison →Folders plus Markdown are flexible and transparent, with the files in your own hands, which suits authors who like to define their own formats; the drawback is that once there are many files, which file is the latest and which other places need syncing after one edit are all yours to remember. Character cards plus story archives fix each character and setting as a card and organize states automatically after a chapter is written, which suits authors who don’t want to maintain files. Maliang has both: characters and settings are cards, the master outline is stored separately, and style samples and book-opening material are documents.
Read the comparison →After signing up, create a book or import existing chapters, tell the AI which chapter to continue, then look at what it wrote, what it changed and what it recorded in the archive, and decide whether to move your long novel over.