produce-long-form-novel

Produce long-form Chinese novels through staged planning, drafting, and continuity tracking.

85|12|Updated Jul 16, 2026
One-click install
npx skills add https://github.com/ExplosiveCoderflome/ani-book-skill --skill produce-long-form-novel-explosivecoderflome
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: produce-long-form-novel
Source: https://github.com/ExplosiveCoderflome/ani-book-skill
Command: npx skills add https://github.com/ExplosiveCoderflome/ani-book-skill --skill produce-long-form-novel-explosivecoderflome

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires PyYAML, and includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve? Long-form fiction projects lose coherence over dozens of chapters: facts drift, foreshadowing is forgotten, and context windows overflow. This Skill turns a vague idea into a managed novel-production workflow with durable continuity state, chapter contracts, and recoverable next steps. ## Core Features & Use Cases - Staged production chain: Move from idea to novel brief, story bible, volume plans, chapter contracts, complete drafts, humanization passes, and review, with only accepted chapters entering long-term memory. - Continuity and cross-book assets: Store facts, foreshadowing, resources, and relationships in authoritative YAML with rebuildable SQLite indexes, plus a cross-book asset graph for shared-IP canon with fork/sync governance. - Research and analysis routes: Analyze public chart metadata for genre trends and deconstruct authorized reference novels into evidence-backed notes and reusable mechanism cards. - Use Case: A writer with only a rough premise invokes the Skill, receives five opening seeds, picks a direction, and then produces and stabilizes one chapter at a time while the workspace tracks every fact and the next safe step. ## Quick Start Use $produce-long-form-novel to plan a long novel from my idea and show me the next production step.

Frequently Asked Questions about produce-long-form-novel

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I start writing a long novel with Codex?

Invoke the Skill with a prompt like "Use $produce-long-form-novel to plan a long novel from my idea." It progressively confirms reader channel and reward, generates two brief previews, then builds the story engine, volume plans, and chapters one at a time.

How does the Skill keep story facts consistent across many chapters?

It stores facts, foreshadowing, resources, and relationships in authoritative YAML files with stable IDs and chapter evidence, using SQLite only as a rebuildable index. Only accepted chapters enter durable memory, and each new chapter gets a minimal context package.

Can I continue an existing novel workspace after an interruption?

Yes. Run novelctl.py status or next on the workspace, or ask the Skill to continue novels/<name>/. It reads the recovery record and state index, validates user edits, and resumes from the next safe production step without rewriting protected prose.

Does the Skill support analyzing other novels or genre trends?

Yes. It deconstructs authorized reference texts into evidence-backed notes and mechanism cards under analyses/, and aggregates public chart metadata into trend reports and opportunity cards under trends/. It never reads prose from charts or bypasses access controls.

What are the limitations of this novel-writing workflow?

It is a Codex-native process contract, not an app runtime: it adds no model SDKs, web APIs, or databases. It produces one chapter at a time rather than whole books in one pass, and chart analysis uses metadata only, never novel prose.