Skip to content

zenstory-ai/oh-story-claudecode / skills/story-review

story-review

Verified

多视角对抗式审查。full/lean 模式在已部署 reviewer agents 时并行 spawn;缺失/异常 agents 或 spawn 失败时自动降级 solo,参考文件不可读时使用内置 rubric fallback。触发方式:/story-review、/审查、「审查一下」「帮我审一下」。

references/solo.md

View on GitHub
# story-review:solo 模式

当前会话直接审查,不 spawn Agent。先按 Phase 1 第 4 步加载通用 rubric 并叠加平台表;文件不可读用内置基准包。

逐项完成基础检查:
1. 格式合规性检查:按戏剧单元/画面分段而非机械按字数切分;段首建立主语、段中用代词或省略(连续无必要重复同一角色名才算主语过密);无段间空行;对话独立成行;正文评价台词、题字等所用的具体字数须统计正确且有叙事必要。
2. 设定一致性 grep(角色名、属性、关键设定、伏笔关键词)+ 推理型一致性检查(规则边界、设定层级、跨章因果链、可滥用漏洞、代价一致性)。
3. AI 味按表达、用笔、故事三层检查:完整读 `story-review/references/expression.md`、`story-review/references/narrative-focus.md`、`story-review/references/story-logic.md` 及 `story-review/references/banned-words.md`(不可读时用 SKILL.md 内置三层去味 fallback 速查并在「没法判断的地方」写明诊断受限)。每项报告原文位置、哪里影响读感、建议与依据,只报告不落改。
4. 按通用 rubric 加叠加的平台表评估内容。
5. 长篇工程里,正文证据显示 `追踪/` 的事实错了或漏了(如伏笔过期、开放钩子漏登记),不改 `追踪/`,把要修的事实写进「需要你决定」,提示作者装好审稿助手后用 `/story-review lean` 修,或在 `/story-long-write` 改到该章时一并提交。
6. 按统一 Findings Schema 整理问题,再按下方模板写给作者。

### solo 模式输出格式

<!-- author-report -->
```md
=== 《{书名}》{审查范围}审查(单人审)===
这次怎么审的:我一个人审{;原因一句话,如"审稿助手还没装,运行 /story-setup 后可以四个视角审"},按通用网文标准{,加上番茄长篇 | 起点长篇 | 短篇平台的要求}。

总体判断:{可以发 | 改完下面几处再发 | 这一章需要重写}——{一句话理由}

## 格式与设定
- 分段与人名节奏:{没问题 | 第N章第M段起……}
- 空行与对话换行:{没问题 | ……}
- 具体数字:{没问题 | 第N章「……」对不上或没必要}
- 前后矛盾:{没发现 | 第N章「……」与第M章「……」矛盾}
- 因果和规则漏洞:{没发现 | ……}

## 必须改({n} 处)
1. 第{N}章「{原文引用}」
   问题:{读者会怎么想、哪里读不通}
   建议:{具体改法}

## 建议改({n} 处)
{同上格式}

## 可以不改({n} 处)
{一行一条}

## 需要你决定
{问题 + 选项 + 我的建议;没有就写"无"}

## 没法判断的地方
{缺哪份设定或大纲导致没法核对、需要人工查证的外部事实;没有就写"无"}

## 下一批接着核对
{仅分批审查填写;否则删掉本节}

下一步:{一句话}
```

---
reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis.Stars: 472kUpdated Claude CodeCodexVerifiedInstall
diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness.Stars: 323kUpdated Claude CodeCodexVerifiedInstall
dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependenciesStars: 323kUpdated Claude CodeCodexVerifiedInstall
executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is availableStars: 323kUpdated Claude CodeCodexVerifiedInstall