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」。
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 18.9 KB |
| references/agent-prompts.md | markdown | 13.4 KB |
| references/banned-words.md | markdown | 10.9 KB |
| references/batch-review.md | markdown | 2.3 KB |
| references/expression.md | markdown | 9.6 KB |
| references/narrative-focus.md | markdown | 6.7 KB |
| references/quality-rubric.md | markdown | 5.1 KB |
| references/review-tracking.md | markdown | 2.2 KB |
| references/rubrics/fanqie.md | markdown | 1012 B |
| references/rubrics/qidian.md | markdown | 1.1 KB |
| references/rubrics/zhihu.md | markdown | 1.4 KB |
| references/solo.md | markdown | 2.8 KB |
| references/story-logic.md | markdown | 5.9 KB |
| references/style-resolution.md | markdown | 3.0 KB |
| references/tracking-initialization.md | markdown | 1.6 KB |
| references/tracking-transaction.md | markdown | 8.8 KB |
| scripts/check-ai-patterns.jsexecutable | script | 79.8 KB |
| scripts/check-degeneration.js | script | 15.4 KB |
| scripts/normalize-punctuation.jsexecutable | script | 19.0 KB |
| scripts/style-whitelist.js | script | 1.1 KB |
| scripts/tracking_commit.jsexecutable | script | 84.7 KB |
| scripts/wordcount_core.jsexecutable | script | 20.3 KB |
references/review-tracking.md
View on GitHub# story-review:审查收尾的追踪文件维护
追踪维护通过本 skill 的 `scripts/tracking_commit.js` 提交事务;[tracking-transaction.md](tracking-transaction.md) 只在报错或字段拿不准时读。**full / lean 模式只允许通过该工具修改 `追踪/`;solo 模式不修改任何 `追踪/` 文件。**不得直接 Edit/Write/追加 `伏笔.md`、角色快照、时间线视图、摘要或 `上下文.md`。
1. **先检查状态**:执行 `tracking_commit.js check --project {项目根}`,确认 `_tracking-state.json` 与全部自动生成的追踪文件一致。失败时按 tracking-transaction.md 的错误类型处理;提交后追踪文件不一致,用新的同章 revision 事务重建,不重跑已清理的旧输入,也不手改 Markdown。
2. **判定是否需要修订**:只有正文证据表明现有追踪事实错误或缺失时才维护。过期伏笔、漏登记开放钩子、角色当前状态、客观时间线、读者认知都归入其证据所在章的 `mode=revision` 事务。审查意见与未来写作建议不进事实追踪。
3. **构造完整同章事务**:先跑 `tracking_commit.js draft --project {项目根} --chapter {N}`,草稿已按该章现有记录预填;保留该章原记录中仍成立的内容,只修改有正文证据的变化;核心角色变化同时提交截至当前最后已写章的完整 `character_snapshots`。伏笔对同一 ID `upsert` 当前状态,不增加重复行;时间线同时提交客观事实、读者当前认知和实际揭示状态。
4. **提交并复检**:执行 `tracking_commit.js commit --project {项目根} --input {事务文件}`,再执行 `tracking_commit.js check --project {项目根}`。确认逐章记录规范且未超限、`上下文.md` 恰好固定 7 栏且 ≤12288 字节、作者/读者时间线及全部自动生成的追踪文件与 state 一致。
5. **正文改过时的出口**:正文改过的修订,commit 会按现正文重新计数。只有报「字数出了作者范围」时不重试、不改字数记录,那要作者定长度。把这章要修的追踪事实写进报告「需要你决定」,请作者在 `/story-long-write` 说「修改第N章」,并告诉它正文已经改过、要重新提交追踪。
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| 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: 472k | Updated | Claude CodeCodex | Verified | Install |
| 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: 323k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 323k | Updated | Claude CodeCodex | Verified | Install |
| 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 available | Stars: 323k | Updated | Claude CodeCodex | Verified | Install |