AI Agent Hub
Back to skills
🏠

Tomato Novel Writing Assistant

Life Service Updated 2026.08.30

Paste the following prompt into your AI chat to install this skill:

Please install @org-1iwqkwdi/tomato-novel according to https://skillhub.cn/install/skillhub.md.

About this skill

What problem it solves

The Tomato Novel Writing Assistant targets long-form web-novel writing issues such as topic drift, pacing problems, and over-polished AI-like prose. It turns vague viral direction into concrete variables: emotion tags, genre, protagonist contrast, core conflict, and chapter scale, so the model does not start from an empty prompt and lose control.

How the skill works

  • Six-question setup: selects the dominant emotion, genre, protagonist setup, core contrast, central conflict, and chapter count.
  • Project planning: creates novels/[novel-name]/ and outputs 00-Outline.md, 01-Character-profiles.md, 02-Emotion-curve.md, and 03-Golden-opening.md.
  • Chapter-by-chapter writing: reads the outline, previous hook, and emotion curve, then writes each chapter toward a 2200–2800 word target with opening conflict, scenes, and closing hooks.
  • Quality and length checks: uses scripts/check_chapter_wordcount_tomato.py to validate word count and applies consistency, dialogue, quotable line, and rhythm checks.
  • Stage feedback: after every five chapters, it generates a pacing report and can use completion rate, follow-read rate, and comment keywords to adjust the outline.

Boundaries

It is a Tomato-style web-novel workflow, not a general academic writer, business copywriter, or poetry assistant. Results depend on the user’s tag choices, chapter targets, and external feedback; if the story does not fit high-conflict, fast-paced, or emotion-dense patterns, the pacing rules need manual revision.

Use Cases

  • Draft a 10-chapter Tomato-style web novel by confirming emotion tags, protagonist contrast, core conflict, outline, and opening.
  • Write each chapter using the previous hook and emotion curve, target 2200-2800 words, and end with a suspense hook.
  • After every 5 chapters, summarize pacing, climax density, and comment topics, then adjust later chapters using retention feedback.
  • Check chapter word count, then revise dialogue, quotable lines, and rhythm to reduce AI-like phrasing.

Best For

  • Web novelists who want Tomato-style short or mid-length thrill stories with stable outlines for setup, contrast, and conflict.
  • Serial writers who need to keep each chapter within word limits, ending hooks, and emotion-curve tasks.
  • Author-operators who have completion, follow-read, and comment data and want to retarget later chapters and supporting characters.
  • Novel editors who need to revise AI drafts into more colloquial, conflict-driven prose and run a quality checklist.