novel-master for Openclaw

An advanced engineering framework for generating publication-grade long-form Chinese novels using distributed character state machines and compressed narrative graphs.

bbroot
v2.0.0
Jul 8, 2026
0
381
0

Install & Download

1. ClawHub CLI

The fastest way to install a skill directly from the registry.

npx clawhub@latest install novel-master-v2

2. Manual Installation

Copy the skill folder to one of these locations

Global
~/.openclaw/skills/
Workspace
<project>/skills/

Priority: Workspace > Local > Bundled

3. Prompt Installation

Copy this prompt to OpenClaw to install it automatically.

Help me install novel-master-v2 using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).

Prefer to download?

Get the raw skill files in a ZIP archive.

What is novel-master?

novel-master is a professional-grade creative engineering workspace designed for generating publication-quality Chinese fiction. Moving past the simplistic nature of commercial web fiction, it focuses on precise linguistic economy, deep emotional resonance, and complex literary structures based on classic dramaturgical theories like the Three-Act Structure, Syd Field's paradigms, and traditional Chinese literary criticism like Wen Xin Diao Long.

The core value proposition of novel-master lies in its breakthrough v2 Distributed Creative Architecture. By abandoning the monolithic single-agent context window paradigm, this skill introduces Process-per-Character (PpC) state separation and Node-Compressed Graphs (NCG) to mitigate context drift. It allows users of Openclaw Skills to seamlessly orchestrate multi-million-word long-form epics without narrative breakdown, out-of-character (OOC) behavior, or plot holes.

novel-master Use Cases

  • Creating high-quality, publication-ready long-form Chinese novels or short story collections across diverse genres.
  • Planning complex multi-arc story world-building, structural frameworks, and granular character voice arcs.
  • Writing sequential chapters under rigorous structural constraints while tracing motifs, subplots, and narrative hooks.
  • Tracking and resolving structural plant-and-payoff setups across a multi-million-word text without context bloating.
  • Executing semantic multi-dimensional conflict detection and real-time narrative quality control.

How novel-master Works

  1. Project Initiation: The agent loads foundational literary templates and structures to output a comprehensive Planning Document that standardizes world logic, character voice lines, and thematic arcs.
  2. Outline Architecture: Generates cohesive multi-volume and chapter-level outlines that map out core structural conflicts, perspective shifts, and hook placement.
  3. Granular Execution Layer: Executes single-perspective chapter drafts by drawing isolated parameter packets, character voice matrices, style rules, and unresolved plot tracking arrays.
  4. Distributed Synchronization: Broadcasts chapter facts via a central Plot Bus to character processes, prompting state machine changes while freezing historical snap-shots.
  5. Graph Settle & Audit: Compresses narrative milestones down to highly dense semantic fragments inside a Node-Compressed Graph and pushes them through a strict multi-dimensional validation gate to ensure consistency.

novel-master Setup

To initiate the workspace structure or access advanced script loops inside your configured environment, use the provided Python scripts. Make sure you place your assets within the managed data directory path:

# Initialize a character sub-process with specific cognitive bias profiles
python scripts/char_state.py init "Your_Book_Name" "Character_Name" --bias "bias_id"

# Compress and settle a newly drafted chapter node into the graph
python scripts/ncg.py settle "Your_Book_Name" 5 --text "Chapter content..." --tags "tag1,tag2"

# Check project versioning and enforce validation criteria
python scripts/auditor.py gate "Your_Book_Name" 5

novel-master Data Schema & Taxonomy

The workspace standardizes creative file outputs under ~/.qclaw/workspace/novels/<book_name>/ using the following taxonomy:

Directory / File Format Purpose
settings/world.md Markdown Comprehensive master document covering global systems and setting definitions.
settings/characters/ Markdown Directory containing isolated data sheets tracking active character voice configurations.
outline/arc-outline.md Markdown Multi-volume master outline dictating the structural progression of overarching themes.
tracker/foreshadowing.json JSON Relational array tracking structural plant IDs, target chapters, and closure statuses.
tracker/plot_bus.json JSON Serialized event bus logging absolute sequential structural facts.
state.json JSON High-level index capturing operational book state metrics and unresolved metadata tags.

novel-master Advanced Features

  • Process-per-Character (PpC): Isolates individual character states into separate state machines communicating only over a secure Plot Bus to eliminate cross-contamination.
  • Node-Compressed Graph (NCG): Compresses raw text into hyper-dense semantic data points under 200 words, allowing multi-perspective projections across timelines and causal links.
  • Quantum Audit Gating: Programmatically blocks compilation steps if a multi-dimensional analysis triggers errors in time, location, character naming, or plot hook rules.
  • Cognitive Bias Injection Module: Injects distinct behavioral biases, asymmetric sensory mappings, and communication rules into characters to eliminate homogenous AI generation.
  • Crash Recovery Loop: Reconstitutes corrupted or lost character context tracking arrays on demand by replaying the system's graph state without reprocessing the entire text base.

SKILL.md


Loading

Related Openclaw Skills

METADATA

Github Stars: 0
forks: 0

Featured*