Chinese Novelist
Paste the following prompt into your AI chat to install this skill:
Please follow https://skillhub.cn/install/skillhub.md to install @user_d5c3472c/chinese-novelist.
About this skill
Problem
Long-form novel generation often collapses into outline recitation: thin character motivation, weak chapter tension, mid-plot drift, and hard recovery after interruption. chinese-novelist splits the workflow into questioning, planning, writing, and validation stages, using structured project files to preserve state so the model does not lose settings while writing.
How It Works
It starts by loading user preferences and detecting unfinished projects, allowing continuation from a breakpoint. Then it uses layered questions to define core positioning: genre idea, protagonist setup, and central conflict, followed by world-building, narrative voice, target readers, chapter count, and other specs. After confirmation, it creates a ./chinese-novelist/{timestamp}-{novel-name}/ directory and generates 00-大纲.md, character profiles, and 03-写作计划.json.
The writing phase supports serial, where the main Agent writes chapter by chapter; subagent-parallel, which batches chapters to sub-Agents; and agent-teams, a Claude Code multi-Agent collaboration mode with inter-Agent communication. Before each chapter, the Agent reads the relevant outline section and follows the rules: show rather than tell, drive plot with conflict, and end chapters with hooks. Validation checks completion and word count, rewriting failed chapters up to three times.
Boundaries
This skill fits complete Chinese novel drafts, long-text state management, and batch chapter generation. For highly specialized factual, legal, medical, or compliance-heavy content, human review is still required. Parallel writing can increase throughput, but character consistency and payoff depend on the outline and character profiles.
Use Cases
- Web novel authors expand a genre premise into an outline with core conflict and chapter hooks while preserving protagonist motivation.
- Editors review chapter word counts and completion, then ask for rewrites of weak chapters using the automatic validation and repair workflow.
- Content teams batch multiple first drafts with sub-Agents and coordinate shared state through the writing plan JSON file for longer novels.
- Writers resume interrupted long-form projects by detecting unfinished work and continuing from the breakpoint without losing established settings and character arcs.
Best For
- Novelists writing long-form genre fiction who need stable outlines, character profiles, and chapter drafts.
- Content editors responsible for specs and manuscript quality who need word-count, completion, and chapter checks.
- Engineers experimenting with Claude Code multi-Agent workflows who want parallel writing and state coordination.
- Screenwriters or planners expanding ideas into full Chinese novel drafts while controlling consistency and hooks.
Related Skills
Rewrites MD, PDF, or TXT material into a mobile-friendly WeChat article with title candidates, outline, and self-check.
A conversational workbench for Chinese long-form novel writing, supporting setup, characters, worldbuilding, chapter planning, drafting, and continuity tracking.
Creates 10,000-30,000-word Tomato Novel short stories with trend research, topic selection, outlines, segmented writing, pacing control, and UTF-8 txt export.
An AI editorial master skill that manages the full content creation process from vague ideas to publishable drafts, covering topic selection, writing, review, and publication adaptation.