Heyvhuang/copybook-skills / zh/skills/review-tips
review-tips
Verified验收 · 16 条:PR、代码和计划复查、bug 打分、红队提示词、skill 回测。
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 4.8 KB |
| agents/openai.yaml | config | 202 B |
| references/3-checks-before-a-skill-install.md | markdown | 2.1 KB |
| references/blast-radius-and-4-more-prompts.md | markdown | 5.7 KB |
| references/bug-score-rule-for-astra.md | markdown | 2.4 KB |
| references/claude-api-prompt-audit.md | markdown | 2.8 KB |
| references/code-and-direction-review.md | markdown | 4.4 KB |
| references/fake-busy-check.md | markdown | 2.5 KB |
| references/ios-simulator-ui-bug-check.md | markdown | 2.8 KB |
| references/logging-plan-and-daily-fix-list.md | markdown | 2.9 KB |
| references/nightly-push-review-routine.md | markdown | 3.3 KB |
| references/one-question-before-it-answers.md | markdown | 2.3 KB |
| references/perf-review-the-5-slowest-spots.md | markdown | 2.9 KB |
| references/plan-review-5-methods.md | markdown | 2.7 KB |
| references/skill-backtest-on-codex-sessions.md | markdown | 3.4 KB |
| references/skill-score-by-yes-no-questions.md | markdown | 3.2 KB |
| references/state-machine-check.md | markdown | 3.4 KB |
| references/tailscale-preview-links.md | markdown | 2.2 KB |
SKILL.md
View on GitHub---
name: review-tips
description: "验收 · 16 条:PR、代码和计划复查、bug 打分、红队提示词、skill 回测。"
disable-model-invocation: true
---
# 验收 · 16 条
Copybook 的 16 条「验收」技巧,收藏多的在前。每条要用的文字、放到哪里、最后得到什么,都在 references/ 里它自己的文件中。
- 用户点名技巧时,用点名的那条。用户只说了要做什么时,从适用于实际执行者的技巧里,选最符合用户目标和限制的一条。几条同样合适时,选一条即可。选中后,用一句话说明选了哪条、为什么,再读它的文件照着做。另一条也合适,而且换用它会明显改变做法或结果时,顺带点出它的名字和区别。
- 用户什么都没说:把下面的列表(每条的名字和做什么)给用户看,问用哪条。
- 都不合适就直说,并建议用 copybook,它会在全部 144 条里找。
| 技巧 | 做什么 |
| --- | --- |
| [手机上的 Tailscale 预览链接规则](references/tailscale-preview-links.md) `tailscale-preview-links` | agent 给的链接是 localhost:预览、审查页面、看板都换成 Tailscale 链接,手机上直接打开。 |
| [Astra 审查的 bug 打分规则](references/bug-score-rule-for-astra.md) `bug-score-rule-for-astra` | Astra 总挑几乎不会发生的 bug,Opus 一条条去修:给 bug 打分,分低的记成 issue,代码不动。 |
| [影响范围、红队等 5 条让模型判断的提示词](references/blast-radius-and-4-more-prompts.md) `blast-radius-and-4-more-prompts` | 只说「修这个 bug」只拿到默认答案:影响范围、并行目标、红队计划、真实用户走查、思考预算。 |
| [iOS 模拟器 UI bug 检查,含 iPhone Duo](references/ios-simulator-ui-bug-check.md) `ios-simulator-ui-bug-check` | iOS app 里的 UI 问题:主要页面都点一遍,有 Duo 模拟器就查各形态,建议最小改法。 |
| [性能审查:最慢的 5 处](references/perf-review-the-5-slowest-spots.md) `perf-review-the-5-slowest-spots` | 想让代码变快,多半换来一层缓存:profile 测出的耗时,每处对上的改法,要改的行和预计提速。 |
| [回答前先问一个问题](references/one-question-before-it-answers.md) `one-question-before-it-answers` | 一次问 10 个你就不想答:先摆出没说的假设、缺的信息和常见错误,再只问最可能改变做法的那个。 |
| [skill 回测:过去的 Codex 会话](references/skill-backtest-on-codex-sessions.md) `skill-backtest-on-codex-sessions` | 凭记忆写的 skill 和实际做法对不上:拿过去的 Codex 会话回测,先出 dry run 的 diff,再定期复查。 |
| [代码和方向复查提示词](references/code-and-direction-review.md) `code-and-direction-review` | 代码越来越复杂,或忙了很久不知道通向哪里:一条找出不必存在的代码,一条重新审视方向。 |
| [/claude-api prompt-audit:换 Fable 5.1 后的 skill 审查](references/claude-api-prompt-audit.md) `claude-api-prompt-audit` | skill 里还留着旧模型名和失效命令:官方 skill 审一遍,列出重复约束和过度强调,附 diff。 |
| [瞎忙检查:今天的 git 历史](references/fake-busy-check.md) `fake-busy-check` | 开了一堆会话,回头看什么都没做成:分清今天真合并的和来回改的,一句话定下明天第一件事。 |
| [每晚审查当天推送的 routine](references/nightly-push-review-routine.md) `nightly-push-review-routine` | 今天推到 main 的代码,睡觉时再看一遍:每个 bug 开 draft PR 附先失败的测试,笔记本合上也跑。 |
| [日志方案和每天的待修清单](references/logging-plan-and-daily-fix-list.md) `logging-plan-and-daily-fix-list` | 坏掉的功能、卡住的用户、太慢的步骤:按影响人数排序,附日志原文,最严重的一条已有测试。 |
| [一条消息里的 5 种计划复查方法](references/plan-review-5-methods.md) `plan-review-5-methods` | 计划写完代码还没动:一条消息要第一性原理、对抗性审查、消融测试、奥卡姆剃刀和高内聚低耦合。 |
| [skill 的是非题评分](references/skill-score-by-yes-no-questions.md) `skill-score-by-yes-no-questions` | skill 改了一版有没有变好:10 个真实任务、8 道是非题、不知情的评分 subagent,只留提分的那版。 |
| [装 skill 前的 3 项检查](references/3-checks-before-a-skill-install.md) `3-checks-before-a-skill-install` | 有人在 ClawHub 上放了 314 个恶意 skill:装之前查来源、扫代码、问清权限,不到 5 分钟。 |
| [状态机检查:错误的转换标红](references/state-machine-check.md) `state-machine-check` | 功能状态间的 bug:从 UI 到服务端的每个状态画在一页 HTML 上,附编号清单,写明文件、行号和原因。 |
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 |