story-review

Run multi-agent rubric-driven story reviews and generate structured severity reports.

1|Updated May 16, 2026
One-click install
npx skills add https://github.com/LoveIiei/oh-story-opencode --skill story-review-loveiiei
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: story-review
Source: https://github.com/LoveIiei/oh-story-opencode/tree/main/.opencode/skills/story-review
Command: npx skills add https://github.com/LoveIiei/oh-story-opencode --skill story-review-loveiiei

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

It helps you catch story problems early by running a multi-agent review across structure, characters, writing quality, and setting consistency, then consolidating findings into an actionable verdict.

Core Features & Use Cases

  • Multi-agent parallel scrutiny (Full/Lean/Solo): Full mode runs four specialized agents in parallel; Lean mode targets structure and setting consistency; Solo mode performs a single-pass self-check.
  • Evidence-based scoring with severity levels (S1-S4): Each issue is scored and must include clear textual evidence, with blockers clearly separated from optional improvements.
  • Platform-aware rubrics: Loads quality rubrics for common Chinese publishing platforms (e.g., Fanqie/Qidian/Zhihu) or uses the general rubric when unspecified.
  • Aggregated problem list and prioritized fixes: Deduplicates overlapping issues, resolves conflicts with defined agent weights, and outputs a structured report with prioritized modification plans.

Quick Start

Run story-review on your target chapters and ask it to perform a full review for the qidian platform.

Frequently Asked Questions about story-review

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

FAQPage Schema
How do I review a novel chapter for character and setting consistency?

To review novel chapter consistency, you can run a multi-agent story review that evaluates character design, setting consistency, writing quality, and architecture. The Skill scans your book chapters and produces a structured report with severity levels and evidence-based fixes.

Can I check my fiction against Qidian or Fanqie platform standards?

Yes, you can check fiction against platform-specific standards like Qidian, Fanqie, or Zhihu. The Skill loads platform-aware quality rubrics during the story review to ensure your chapters meet the target platform's publishing requirements.

What is the best way to audit a long-form network novel for plot holes?

The best way to audit long-form network novels is running a full multi-agent story review. Four specialized agents scrutinize structure and setting in parallel, aggregating plot holes into a deduplicated, prioritized modification plan with clear textual evidence.

How do I get an actionable revision verdict for my fiction manuscript?

You get an actionable revision verdict by running a rubric-driven story review. It resolves overlapping issues and agent conflicts using defined weights, outputting a structured report that separates critical blockers from optional improvements.

Does the story review process work for quick single-chapter checks?

Yes, the story review process works for quick single-chapter checks using Solo mode. It performs a single-pass self-check for fast revisions, while Lean mode targets structure and setting, and Full mode runs four agents for complete audits.

Why should I use a multi-agent review instead of manually editing my web novel?

A multi-agent review prevents human oversight by running specialized agents in parallel for writing quality and character consistency. It enforces evidence-based scoring with S1-S4 severity levels, ensuring objective, structured problem resolution over manual editing.