story-review
story-review coordinates multi-perspective critique of story drafts, identifying structural, character, prose, and consistency problems with executable revision guidance. It operates in full, lean, or solo modes—spawning specialized agents when available, or degrading gracefully to solo analysis when agents are unavailable or misconfigured. Includes platform-specific rubrics for Fanqie, Qidian, and Zhihu, with embedded fallback standards for universal web fiction assessment.
story-review finds structural, character, prose, and consistency problems in story drafts across multiple review modes.
AI-generated summary based on this skill's SKILL.md
Install
worldwonderer/oh-story-claudecode/story-review · repository language: JavaScript
git clone https://github.com/worldwonderer/oh-story-claudecode
cp -r oh-story-claudecode/skills/story-review ~/.claude/skills/story-reviewnpx skillfed install worldwonderer/oh-story-claudecode/story-reviewFrequently asked questions
AI-generated answers based on this skill's SKILL.md and metadata
What is story-review and what can it do for my draft?
story-review coordinates multi-perspective critique of story drafts, identifying structural, character, prose, and consistency problems with executable revision guidance. It analyzes your narrative across multiple angles—spawning specialized agents when available—to catch issues a single pass might miss. The skill includes platform-specific rubrics for Fanqie, Qidian, and Zhihu, plus universal web fiction standards.
Can story-review detect AI-generated writing patterns?
Yes. story-review includes detection for AI-generated writing patterns and enforces natural prose standards as part of its quality assessment. This helps ensure your story maintains authentic voice and human-like narrative flow rather than synthetic phrasing that readers recognize as machine-generated.
What happens if story-review's reviewer agents are unavailable?
story-review automatically degrades to solo review mode when parallel reviewer agents are unavailable or misconfigured. This fallback ensures you still receive comprehensive analysis—structural, character, prose, and consistency feedback—even when the multi-agent system cannot spawn, so your review never fails completely.
How does story-review validate stories against quality standards?
story-review validates your story against platform-specific quality rubrics tailored to Fanqie, Qidian, and Zhihu, as well as universal web fiction assessment standards. It checks narrative structure, character consistency, prose quality, and overall coherence to ensure your draft meets the expectations of your target audience and platform.
Does story-review work in different operational modes?
story-review operates in full, lean, or solo modes. Full mode spawns specialized agents for multi-angle critique; lean mode uses fewer agents for faster turnaround; solo mode provides single-pass analysis when agents are unavailable. You can choose the mode that fits your timeline and available resources.
What structural and character issues can story-review identify?
story-review finds structural problems like plot holes, pacing issues, and narrative architecture flaws, plus character consistency problems including voice drift, motivation gaps, and relationship contradictions. It also flags prose quality issues and overall coherence gaps, providing specific revision guidance for each problem detected.
SKILL.md
rendered from the published skill — quoted content, verbatim
story-review：多视角对抗式审查
你是审查协调器。你的职责是找出小说文本中的结构、角色、文字、设定问题，并给出可执行修改建议。
执行铁律：审查是找问题，不是验证正确性。
Review Mode 选择
/story-review或/story-review full→ 优先 spawn 全部 4 个 Agent；如果当前已经在子代理内，核心 Agent 未部署/异常，或 spawn 失败，自动降级为 solo。/story-review lean→ 优先 spawnstory-architect+consistency-checker；如果当前已经在子代理内，任一所需 Agent 未部署/异常，或 spawn 失败，自动降级为 solo。/story-review solo→ 不 spawn Agent，由当前会话执行基础审查。- 未指定 → 默认 full，并在报告里写明最终实际执行模式。
> AI味 / 文字自然度这一维度只有 narrative-writer 审，仅 full 模式覆盖。lean 只 spawn story-architect + consistency-checker，审的是结构与设定一致性，不含文字自然度审查；要审文字层是否像人写，用 full。
Phase 0：预检与降级（必须先执行）
- 确定请求模式：解析用户输入中的
full、lean、solo；未指定时目标模式为full。 - 确认是否允许 spawn：如果当前已经在子代理/Agent 内执行，不再递归 spawn，直接降级为
solo。 - 识别 ZCode 能力边界：如果当前运行于 ZCode 且项目使用
.zcode/，ZCode 3.3.4 不执行项目/plugin custom agents；不要因为磁盘上存在其他端的 agent 文件就尝试同名 spawn，直接降级solo
(truncated - see the full file via the links below)
Read as markdown · JSON record · Browse the source repository
File tree — 14 files
skills/story-review/SKILL.md
skills/story-review/references/anti-ai-writing.md
skills/story-review/references/banned-words.md
skills/story-review/references/character-relations.md
skills/story-review/references/dialogue-mastery.md
skills/story-review/references/plot-core-methods.md
skills/story-review/references/quality-checklist.md
skills/story-review/references/quality-rubric.md
skills/story-review/references/rubrics/fanqie.md
skills/story-review/references/rubrics/qidian.md
skills/story-review/references/rubrics/zhihu.md
skills/story-review/scripts/check-ai-patterns.js
skills/story-review/scripts/check-degeneration.js
skills/story-review/scripts/normalize-punctuation.js