Setup Guide

Everything you need to get bookmoth running,
from first install to first chapter.

On this page
What is bookmoth? Installing bookmoth Setting up Anthropic (Claude) Setting up OpenRouter Setting up Ollama (local) Setting up LM Studio (local) Using Hybrid mode Advanced model routing Your first project Writing a series The Editor The Bible (cast & continuity) Windows notes Getting help

What is bookmoth?

A desktop writing app for novels and long-form non-fiction. You bring your project, your voice, and your own AI provider. bookmoth learns how you write and keeps everything it produces within your style.

Take the weight out of the first draft with a guided brief and AI-drafted chapters, or import an existing manuscript and use bookmoth for editorial feedback, planning, and revision. Either way, you stay in control with a sidekick that knows what you're trying to do.

bookmoth doesn't include an AI model. You connect one (Anthropic, OpenRouter, Ollama, or LM Studio) and pay the provider directly. A full novel typically costs under $5 in usage. This guide walks you through setup.

Installing bookmoth

Purchase bookmoth from the store. You'll receive a serial key by email. Download the latest version from the download page.

Open the installer. On macOS, open the .dmg and drag bookmoth to Applications. On Windows, run the .exe installer.

Enter your serial key on first launch. This is the key from your purchase email. One key per machine. You can transfer it later from the settings.

Set up your AI provider. bookmoth needs an API connection to generate text. Choose one of the options below.

Updates. When a new version is available, bookmoth will let you know. Download and install it over the top. Your serial key, projects, and all settings carry across automatically.

Setting up Anthropic (Claude)

Anthropic's Claude models produce the highest quality output. This is the recommended setup for serious drafting.

Go to console.anthropic.com and create an account if you don't have one.

Navigate to API Keys and create a new key. Copy it.

In bookmoth, click API / LLM in the top right. Select Cloud mode.

Paste your API key and save.

Cost. A 3,500-word chapter costs roughly 10 cents. A full novel typically runs $3 to $5 in total API usage. You control your spend through your Anthropic dashboard.

Setting up OpenRouter

OpenRouter gives you access to multiple AI models (Claude, GPT, Gemini, open-source models) through a single API key. Often cheaper than direct Anthropic access.

Go to openrouter.ai and create an account.

Navigate to Keys and create a new API key. Copy it.

In bookmoth, click API / LLM. Select Custom Endpoint.

Enter the endpoint URL:

https://openrouter.ai/api/v1/chat/completions

Enter the model name in the model field. Use the OpenRouter model ID, e.g. zhipu-ai/glm-5, anthropic/claude-sonnet-4, or deepseek/deepseek-r1. You can find model IDs on the OpenRouter models page.

Paste your OpenRouter API key in the key field and save.

Which model? For fiction writing, zhipu-ai/glm-5 offers excellent voice fidelity at a fraction of the cost of Opus. See the model field in bookmoth's API settings to set your preferred model per project.

Setting up Ollama (local, offline)

Run AI models entirely on your machine. No internet needed, no API costs. Requires a reasonably powerful computer (16GB+ RAM recommended).

Install Ollama from ollama.com.

Open Terminal (macOS) or Command Prompt (Windows) and pull a model:

ollama pull llama3

Start the Ollama server if it's not already running:

ollama serve

In bookmoth, click API / LLM. Select Custom Endpoint.

Enter the endpoint URL:

http://localhost:11434/v1/chat/completions

Enter your model name exactly as it appears in Ollama. To check, run ollama list in terminal and use the name from the left column (e.g. llama3, mistral, qwen2.5).

Leave the API key field blank. Save.

Which model? bookmoth's tasks vary in complexity. Writing profile generation, chapter drafting, and editor chat are heavy calls that need a capable model, 30B parameters or larger recommended. Lighter tasks like scene planning and inline edits work fine with smaller, faster models. If you're getting errors or garbled output on profile generation, try a larger model. Sketch mode is designed for smaller models.
Timeouts. Local models can be slow on large prompts. bookmoth gives Ollama calls 30 minutes by default, so it won't cut out on you. You can adjust this in the advanced model routing panel. A cancel button is always visible during generation if you want to stop early.

Setting up LM Studio (local, offline)

LM Studio provides a visual interface for downloading and running local models. Good if you prefer a GUI over the command line.

Download LM Studio from lmstudio.ai.

Search for and download a model inside LM Studio (the Discover tab).

Go to the Developer tab and click Start Server.

In bookmoth, click API / LLM. Select Custom Endpoint.

Enter the endpoint URL:

http://localhost:1234/v1/chat/completions

Enter the model name as shown in LM Studio's Developer tab (it appears at the top of the server panel once loaded).

Leave the API key field blank. Save.

Model size matters. Writing profile generation and chapter drafting are heavy calls. If you're getting errors, try a larger model (30B+ parameters). Smaller or "flash" variants work well for lighter tasks and Sketch mode but may not handle the bigger prompts. bookmoth gives LM Studio calls 30 minutes by default, and you can cancel at any time during generation.

Using Hybrid mode

Hybrid mode splits your API calls between Anthropic and your Custom Endpoint based on how demanding the task is. Quality-critical work stays on Claude. Everything else runs through your cheaper or local provider.

Heavy tasks always use Anthropic (Claude): chapter drafting, writing profile generation, editor chat, polish and voice revision. These are the calls where output quality matters most.

Medium and light tasks try your Custom Endpoint first: scene generation, chapter summaries, bible extraction, inline edits, slash commands. If the custom endpoint fails or is unreachable, bookmoth silently falls back to Anthropic. You never see an error.

Enter your Anthropic API key in the top section of API / LLM settings.

Select a provider tab (OpenRouter, Ollama, LM Studio, or Custom) and enter your model name and API key if needed.

Click Test connection on both Anthropic and your custom endpoint to verify they're working.

Select Hybrid mode.

Cost savings. Roughly 70% of bookmoth's API calls are medium or light tasks. Running those through a local model or a cheaper provider like OpenRouter can significantly reduce your Anthropic bill while keeping your chapter drafts on Claude.

Advanced model routing

For most users, the standard routing modes (Cloud, Hybrid, Custom) are all you need. Advanced routing is for writers who want finer control over which model handles which type of task.

In API / LLM settings, click Advanced model routing below the mode selector. This opens a panel where you can assign a different model to each of bookmoth's three task tiers:

Heavy tasks: chapter drafting, writing profile generation, editor chat, polish pass. These prompts are long, complex, and quality-critical. Use a frontier API model or a local model with 70B+ parameters. Smaller models will produce poor output or fail entirely.

Medium tasks: scene generation, chapter summaries, bible extraction, brief drafting. Structured output that needs good reasoning but not frontier-level prose. 14B+ parameters recommended.

Light tasks: inline edits, slash commands, quick classification. Speed matters more than depth. 7B+ parameters is fine.

Leave any tier blank and it uses your default model from the main settings. This means you only need to fill in the tiers you want to override.

A common setup for local users running Ollama: assign a large model like qwen2.5:72b to Heavy, a mid-size model like qwen2.5:14b to Medium, and a fast model like llama3:8b to Light. Same endpoint, three different models, each matched to the task.

The advanced panel also includes a timeout setting. bookmoth sets sensible defaults per provider (2 minutes for OpenRouter, 30 minutes for local models), but you can override it here if you need more or less time.

Fallback. If a model fails or times out, bookmoth will retry the call using your Anthropic key (if one is configured). If no Anthropic key is set, you'll see the error directly.

Your first project

Once your AI provider is connected, you're ready to start.

Let the Editor guide you. You can fill in the brief, series context, and chapter plan manually, but the Editor is designed to walk you through it. Just have a conversation about your project and it builds everything for you. It's faster and usually produces richer results than typing into fields directly.

Create a new project from the sidebar. Give it a name.

Talk to the Editor. The Brief tab opens a guided conversation. Tell the Editor about your novel: premise, characters, tone, world. The more you put in here, the better your chapters will be.

Add your writing examples. Click Writing Examples and paste in a few pages of your own prose. bookmoth analyses this to build your voice profile.

Generate your writing profile. Once you have a brief (at least 500 words) and writing examples, bookmoth creates a long-form voice profile that governs every draft.

Plan your chapters. Use the Chapter Plan tab to create your structure, or let the Editor help you map it out.

Write. Select a chapter in the Manuscript tab and click Write Chapter. Your first draft will be governed by your brief and your voice profile.

Bringing an existing manuscript? You can import a .md, .txt, or .docx file. Make sure each chapter starts with "Chapter 1", "Chapter 2" etc. on its own line. Markdown headings (## Chapter 1) work too. Coming from Scrivener? Compile to Plain Text using the default preset.

Writing a series

If you're writing the second (or third, or fourth) book in a series, bookmoth can carry context forward from previous books. The easiest way to set this up is through the Editor.

When you start a new project, tell the Editor it's a sequel. Something like "this is book 2 in a series" or "I'm continuing from my previous novel" is enough. The Editor will ask you about what happened in the previous book, the world you've established, any continuing plot threads, and which characters are returning. It populates the series context fields and returning characters for you automatically.

You don't need to fill anything in manually. Just have the conversation and the Editor does the rest.

If you prefer to do it yourself, expand the Series Context card in the Brief tab and enable it. You'll see three text fields:

Starting situation: where things left off at the end of the last book. What happened, where are the characters now, what's unresolved.

World and rules: anything established in previous books that still applies. Magic systems, political structures, geography, social dynamics.

Continuing threads: plot threads, relationship dynamics, unresolved tensions, or promises to the reader that carry across books.

Returning characters are managed in the Bible tab's Characters section. Click "+ Add returning character" to add characters from previous books with their descriptions. These appear in the cast grid with a "Series" badge and are visible to the AI when drafting.

Each book is its own project. Create a new project for each book. The series context carries forward what the AI needs to know. Your writing examples and voice profile can be the same across books for consistency. The Editor is the fastest way to get set up: just tell it about the previous book and it handles the rest.

The Editor

The Editor is the chat panel on the right side of the Brief tab. It's a guided conversation that helps you build and refine your project. Think of it as a developmental editor on call.

The Editor can update your brief, style guide, chapter plan, dos and don'ts, constraints, and series context. When it proposes changes, it will summarise what it wants to do and ask for confirmation before applying anything.

You can use the Editor at any stage of your project:

Early on: describe your premise, characters, tone. The Editor asks follow-up questions to deepen your brief. The richer this gets, the better everything downstream works.

Chapter planning: ask the Editor to map out your chapter structure. It outputs chapter titles, summaries, and act assignments directly into your chapter plan.

Mid-project: need to change a character's arc, adjust tone, or restructure? Tell the Editor. It makes surgical edits to the relevant documents without rewriting everything else.

After drafting: use "Ask the Editor" on any chapter for editorial feedback: pacing, voice, structure, dialogue, whatever you want a second opinion on.

The Editor sees everything. Your brief, writing profile, chapter plan, and series context are all visible to the Editor in every conversation. You don't need to re-explain context between sessions.

The Bible (cast & continuity)

The Bible tab tracks your characters and continuity across chapters. It has two sections: Characters (the cast grid) and Codex (manual reference entries).

Characters are populated automatically. After you draft a chapter, bookmoth parses the text and extracts character appearances, descriptions, and details into the cast grid. Each character card shows which chapters they appear in and key details from each appearance. This builds up over time as you draft more chapters.

You don't need to set this up manually. Draft your chapters and the cast grid fills itself in.

Codex entries are manual. Use these for world-building details, location descriptions, magic systems, timelines, or anything you want to reference but that doesn't fit into a character card. Click "Add to Codex" and fill in a name and description.

Bible generation is in beta. Character extraction is automated and works well for most prose, but it's not perfect. It may occasionally miss a character or pull in a name that isn't a character. You can always add, edit, or remove entries manually.

Windows notes

bookmoth is independent software, not distributed through the Microsoft Store. Windows will show a SmartScreen prompt the first time you run it. This is normal for any new, independently distributed application.

To install: click More info on the SmartScreen prompt, then click Run anyway.

bookmoth is safe, built on open-source infrastructure, and stores everything locally on your machine.

Getting help

If you run into anything, email support. It goes directly to Stu, the founder. You'll get a real answer, usually the same day.