Token导航 LogoToken导航

Agent Skill · 开发

copy-editing(复制编辑)

copy-editing 用于处理 GitHub 仓库、Issue、Pull Request 和代码协作信息,适合在 Codex、Claude、Cursor、Gemini CLI 中需要围绕仓库状态、代码变更或协作事项进行整理时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

copy-editing 是开发类 Agent Skill,可用宿主为 Codex、Claude、Cursor、Gemini CLI,通过GitHub 仓库获取,权限为只读;许可证信息还不明确,第三方扫描已通过,按安装量在开发类的 21,746 个 Skill 里排第 11,768。

总安装 1,592周安装 67GitHub Stars 103下载 557核验 2026-05-01

安装说明

本站只整理中文说明,不托管安装包,也不代用户安装。
GitHub
许可证
unknown

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:copy-editing(复制编辑)
来源仓库:https://github.com/borghei/claude-skills
仓库路径:skills/copy-editing
安装命令:
npx skills add https://github.com/borghei/claude-skills --skill copy-editing
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。该命令会通过 npx skills 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

skills.shnpx skills
npx skills add https://github.com/borghei/claude-skills --skill copy-editing

这个 Skill 是做什么的

开发类 Skill 面向写代码、改代码和排查问题,一般会读写仓库文件,部分还会执行构建或测试命令。

copy-editing 通过系统性编辑流程提升文案清晰度、语气一致性与转化效果。适合在 Codex、Claude、Cursor、Gemini CLI 中进行七轮专项校对与事实核查。提供常见问题修复指南与风格标准化工具,强化品牌声音与利益点表达。

安装命令:npx skills add https://github.com/borghei/claude-skills --skill copy-editing。使用前应确认权限范围、维护状态及是否会触发文本修改或输出替换操作。

来源仓库内路径:skills/copy-editing

上手步骤

4 步
  1. 1

    确认自己的宿主工具

    公开资料标注支持 Codex、Claude、Cursor、Gemini CLI;先确认自己用的是其中之一,再按对应工具的 Skills 目录放置。

  2. 2

    拿到安装命令

    可以直接复制的命令是 npx skills add https://github.com/borghei/claude-skills --skill copy-editing;命令里的占位符换成自己的路径或密钥再执行。

  3. 3

    安装后重启宿主

    命令执行完要完全退出宿主工具再打开,让新装的 Skill 被重新扫描到,否则列表里可能看不到。

  4. 4

    按最小权限试用

    它标注为只读,先用只读任务试一次,确认行为符合预期再扩大使用范围。

常见问题

6 条

copy-editing 怎么安装?

在终端执行 npx skills add https://github.com/borghei/claude-skills --skill copy-editing。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

copy-editing 支持哪些 AI 工具?

公开资料标注支持 Codex、Claude、Cursor、Gemini CLI。其他工具只要能读取本地 Skill 目录,一般也能手动放进去,但要自己确认触发方式是否一致。

copy-editing 需要什么权限?

权限级别是只读。该 Skill 主要提供规则、说明或参考内容,本身偏只读;真正读写文件、联网或执行命令仍取决于宿主 Agent 的任务。

copy-editing 是免费的吗?

公开资料没有标注明确许可证,收费和商用限制都无法确认。正式使用前请到来源仓库核对授权说明。

copy-editing 用的人多吗?

公开数据是累计安装 1,592 次,近一周安装 67 次,下载 557 次,GitHub 103 星。这些数字来自第三方目录统计,只能当作热度参考,不代表质量或安全性。

copy-editing 安全吗?

第三方扫描结果为通过,但这只覆盖已知的静态检查项,不能替代自己读一遍源码和安装命令。

值得选的地方和要先想清楚的地方

值得选的地方

  • 同时适配 4 种工具,换客户端不用重新找替代品。
  • 第三方扫描已通过,常见的高风险行为没有命中。
  • 标注为只读权限,试用阶段不容易误改本地文件。

先想清楚的地方

  • 许可证没有明确标注,能不能商用、能不能再分发都不确定。
  • 第三方目录整理,版本更新和行为细节以来源仓库为准。

适不适合你

更适合

  • 已经在用 Codex,想直接扩展它能力的人。
  • 需要处理开发类任务、又不想每次都手写提示词的人。
  • 只想先试试效果、不希望它改动本地文件的人。

不太适合

  • 需要明确授权才能用到公司项目里的场景。
  • 需要厂商级技术支持和 SLA 的生产系统。

开发类 Skill 对比

6 个同类
Skill总安装GitHub 星权限许可证扫描
skill-vetter

技能审查员 · OpenClaw

8,779,018986操作浏览器MIT-0通过
ontology

本体论 · OpenClaw

6,636,960576只读MIT-0通过
entra-app-registration

内部应用程序注册 · Codex、Claude

6,631,872764接触敏感数据未明确通过
entra-app-registration

内部应用程序注册 · Codex、Claude

3,757,680201接触敏感数据未明确通过
nano-banana-pro

纳米香蕉专业版 · OpenClaw

3,526,898364接触敏感数据MIT-0有提醒
copy-editing

复制编辑 · 本页

1,592103只读未明确通过

按安装量取同分类前几名;数字来自第三方目录统计,只作热度参考,不代表质量结论。

SKILL.md 原文

Copy Editing

Systematic copy improvement through focused editorial passes that enhance clarity, voice, proof, and conversion impact.


Table of Contents


Keywords

copy editing, editorial review, copy feedback, proofreading, content polishing, copy sweep, editorial standards, style consistency, grammar check, fact-checking, clarity editing, voice consistency, benefit framing, proof validation, specificity, conversion copy editing, marketing copy review, copy quality


Quick Start

Full Copy Review (Seven Sweeps)

  1. Read through once without editing to understand the whole piece
  2. Sweep 1 — Clarity: flag confusing sentences, unclear references, jargon
  3. Sweep 2 — Voice and Tone: flag shifts in formality, personality inconsistencies
  4. Sweep 3 — So What: flag features without benefits, claims without consequences
  5. Sweep 4 — Prove It: flag unsubstantiated claims, missing social proof
  6. Sweep 5 — Specificity: flag vague language, round numbers, generic statements
  7. Sweep 6 — Heightened Emotion: strengthen pain points, aspirations, urgency
  8. Sweep 7 — Zero Risk: remove barriers near CTAs, add trust signals

Quick Copy Pass

  1. Cut filler words (very, really, just, actually, basically)
  2. Replace weak verbs (utilize > use, facilitate > help, leverage > use)
  3. Fix passive voice (reports are generated > we generate reports)
  4. Verify one idea per sentence, one topic per paragraph
  5. Check CTA for action orientation

The Seven Sweeps Framework

Edit through seven sequential passes. Each focuses on one dimension. After each sweep, verify previous sweeps are not compromised.

Sweep 1: Clarity

Focus: Can the reader understand what you are saying on the first read?

What to check:

  • Sentences trying to say too much (split them)
  • Unclear pronoun references ("it" — what is "it"?)
  • Jargon or insider language without explanation
  • Ambiguous statements that could be read two ways
  • Missing context that assumes reader knowledge

Clarity killers to fix:

ProblemFix
Sentence over 30 wordsSplit into two sentences
Abstract languageReplace with concrete example
Buried main pointMove to beginning of paragraph
Three-clause sentenceSimplify to one or two clauses
Undefined acronymSpell out on first use

After this sweep: Confirm the "Rule of One" (one idea per section) and "You Rule" (copy speaks to the reader as "you") are intact.

Sweep 2: Voice and Tone

Focus: Does the copy sound consistent throughout?

What to check:

  • Shifts between formal and casual language
  • Inconsistent brand personality (joking in one paragraph, corporate in the next)
  • Jarring mood changes without transition
  • Word choices that do not match the established voice
  • Mixing "we" and "the company" references

Voice consistency indicators:

ConsistentInconsistent
Same level of contractions throughoutContractions in some sections, full forms in others
Humor style maintainedRandom joke in otherwise serious copy
Same sentence structure patternsShort punchy intro, corporate middle, casual close
Consistent use of "you"Switching between "you," "users," "customers," "one"

After this sweep: Return to Sweep 1 to ensure voice edits did not introduce confusion.

Sweep 3: So What

Focus: Does every claim answer "why should I care?"

The So What test: For every statement, ask "So what?" If the copy does not answer with a deeper benefit, it needs work.

Before (features only)After (feature + benefit)
"Our platform uses AI-powered analytics""Our AI analytics surface insights you would miss manually — so you make better decisions in half the time"
"SOC 2 Type II certified""SOC 2 certified — your security team approves us in days, not months"
"Real-time dashboard""See exactly what is happening right now, not what happened last week"

After this sweep: Return to Sweeps 2 and 1.

Sweep 4: Prove It

Focus: Is every claim backed with evidence?

Types of proof to verify:

Proof TypeStrengthExample
Named testimonialStrong"Sarah Chen, VP Marketing at Stripe: 'Reduced our setup time by 60%'"
Specific statisticStrong"2,847 teams use [Product] daily"
Case study referenceStrong"See how Linear reduced churn by 23% in 90 days"
Third-party validationStrong"Named a Leader in Gartner's Magic Quadrant 2025"
Customer logosMediumRecognizable brand logos with permission
Generic claimWeak — flag it"Customers love us," "Industry-leading"

Common proof gaps to flag:

  • "Trusted by thousands" (which thousands? give a number)
  • "Industry-leading" (according to whom? cite the source)
  • "Best-in-class" (by what measure?)
  • "Customers love us" (show them saying it)
  • Results claims without timeframe or specifics

After this sweep: Return to Sweeps 3, 2, and 1.

Sweep 5: Specificity

Focus: Is the copy concrete enough to be compelling?

Specificity upgrades:

VagueSpecific
"Save time""Save 4 hours every week"
"Many customers""2,847 teams"
"Fast results""Results in 14 days"
"Improve your workflow""Cut reporting time from 4 hours to 15 minutes"
"Great support""Average response time: 2 hours"
"Easy to use""Set up in 10 minutes, no code required"
"Affordable""Starting at $29/month"
"Scalable""Handles 10,000 to 10 million records without slowdown"

Rule: If a claim cannot be made specific, it is probably filler. Cut it or replace it with something verifiable.

After this sweep: Return to Sweeps 4, 3, 2, and 1.

Sweep 6: Heightened Emotion

Focus: Does the copy make the reader feel something?

Emotional dimensions to check:

EmotionWhere to UseTechnique
Pain/frustrationProblem sectionPaint the "before" state vividly
ReliefSolution sectionShow the contrast with current pain
Fear of missing outSocial proof"Teams like yours already use..."
PrideAspiration section"Be the team that..."
ConfidenceCTA area"Join 2,847 teams who already..."
UrgencyNear CTAOnly if genuine (real deadline, limited spots)

Emotion techniques:

  • Paint the "before" state with sensory detail
  • Use micro-stories (1-2 sentences) from customer scenarios
  • Ask questions that prompt self-reflection
  • Reference shared experiences the audience recognizes

After this sweep: Return to Sweeps 5, 4, 3, 2, and 1.

Sweep 7: Zero Risk

Focus: Have we removed every barrier to action?

Friction checklist near CTAs:

  • What happens after clicking is clear (not a mystery)
  • Objections addressed within 2 scrolls of the CTA
  • Trust signals visible (guarantee, certifications, customer count)
  • Next steps are specific ("Start your 14-day free trial" not "Get started")
  • Risk reversals stated explicitly (money-back, no CC, cancel anytime)
  • Privacy concerns addressed if form collects data

After this sweep: Return through all previous sweeps one final time.


Quick-Pass Editing Guide

Word-Level Cuts

Always cut: very, really, extremely, incredibly, quite, rather, somewhat, just, actually, basically, essentially, literally (unless literal), in order to (use "to"), the fact that, it should be noted that, it is important to

Always replace:

WeakStrong
UtilizeUse
ImplementSet up, build, create
LeverageUse, apply
FacilitateHelp, enable
InnovativeNew, original, first-of-its-kind
RobustStrong, thorough, [be specific]
SeamlessSmooth, easy, [be specific]
Cutting-edgeModern, latest, [be specific]
Synergy[Delete or be specific about the collaboration]
Paradigm[Delete or say what actually changed]

Sentence-Level Checks

  • One idea per sentence
  • Vary sentence length (mix 8-word and 20-word sentences)
  • Front-load important information (do not bury the point)
  • Maximum 3 conjunctions per sentence
  • Active voice default (flip passive constructions)

Paragraph-Level Checks

  • One topic per paragraph
  • 2-4 sentences maximum for web copy
  • Strong opening sentence that states the paragraph's point
  • Logical flow between paragraphs
  • White space for scannability

Common Copy Problems and Fixes

ProblemSymptomFix
Wall of featuresList of what it does, no whyAdd "which means..." after each feature
Corporate speak"Leverage synergies to optimize outcomes"Ask "How would a human say this?"
Weak openingStarts with company history or vague statementLead with reader's problem or desired outcome
Buried CTAAsk comes after too much buildupMake CTA obvious, early, and repeated
No proof"Customers love us" with no evidenceAdd specific testimonials, numbers, case references
Generic claims"We help businesses grow"Specify who, how, and by how much
Mixed audiencesTries to speak to everyonePick one audience per page/section
Feature overloadEvery capability listedFocus on 3-5 benefits that matter most
Passive voice"Reports are generated by the system""The system generates reports"
Weasel words"Up to 50% improvement"State the median or typical result with context

Style Consistency Standards

Checklist for Style Consistency

  • Oxford comma: used consistently (or consistently omitted)
  • Contractions: consistent usage throughout
  • Heading case: consistent (sentence case or title case, not mixed)
  • Number style: consistent (spell out 1-9, numerals for 10+, or chosen standard)
  • Date format: consistent (March 9, 2026 or 2026-03-09, not mixed)
  • Brand name: capitalized and formatted consistently
  • Product/feature names: capitalized consistently per brand standards
  • Bulleted lists: consistent punctuation (periods or no periods)
  • Acronyms: spelled out on first use in each document
  • Em dashes, en dashes, hyphens: used correctly and consistently
  • Quotation marks: consistent style (straight or curly)

Common Style Conflicts

DecisionOption AOption BHow to Decide
Oxford commaYesNoPick one, document it, enforce it
Heading capitalizationSentence caseTitle CaseSentence case is modern standard
"Login" vs "Log in"One word (noun)Two words (verb)"Log in" as verb, "Login" as noun/adjective
"Setup" vs "Set up"One word (noun)Two words (verb)Same pattern as login
Ampersand vs "and"&and"and" in prose, "&" only in headings if brand standard

Fact-Checking Protocol

What to Verify

  • All statistics have a named source and year
  • Customer testimonials are attributed to real, named individuals
  • Customer logos are used with permission
  • Competitive claims are accurate and current
  • Product capabilities described are actually available (not roadmap items)
  • Pricing is current and matches the pricing page
  • Certifications and compliance claims are active (SOC 2, GDPR, etc.)
  • Awards and recognition are current year or specified year
  • Integration claims list actual integrations, not aspirational ones
  • Uptime/SLA claims match the actual SLA

Red Flags to Investigate

  • Round numbers without source (sounds made up)
  • Superlatives without qualification ("fastest," "best," "only")
  • Claims that contradict other pages on the same site
  • Screenshots from an older version of the product
  • Competitor comparisons without date (may be outdated)

Editorial Checklist

Pre-Edit

  • Understand the goal of this copy
  • Know the target audience
  • Identify the desired action
  • Read through once without editing

During Edit (Seven Sweeps Summary)

  • Sweep 1: Every sentence is immediately understandable
  • Sweep 2: Voice is consistent throughout
  • Sweep 3: Every feature connects to a benefit
  • Sweep 4: Claims are substantiated with evidence
  • Sweep 5: Vague words replaced with specifics
  • Sweep 6: Copy evokes appropriate emotion
  • Sweep 7: Barriers to action removed near CTAs

Post-Edit

  • No typos or grammatical errors
  • Consistent formatting throughout
  • Core message preserved through all edits
  • All links functional (if applicable)
  • Consistent style applied (see style checklist)

Best Practices

  1. Edit in passes, not all at once — Trying to fix everything in one read misses issues. Each sweep catches what the others miss.
  2. Preserve the author's voice — Good copy editing enhances; it does not replace. Maintain the original voice while improving clarity and impact.
  3. Every edit needs a reason — Never change a word without explaining the principle. "Changed because it is clearer" or "Replaced because the original was vague."
  4. Prioritize by conversion impact — Fix the CTA before fixing a comma. Fix the headline before fixing paragraph 12.
  5. Read aloud — Voice and rhythm problems become obvious when read aloud. If it sounds wrong spoken, it reads wrong too.
  6. Flag what you cannot fix — If a claim needs proof the author must provide, flag it clearly. You can improve phrasing but you cannot invent evidence.
  7. Re-check previous sweeps — Each sweep can introduce issues caught by earlier sweeps. Always go back.
  8. Cut first, add second — Most marketing copy is 20-30% too long. Cut the fat before adding new content.
  9. Get context before editing — A copy edit without knowing the audience, goal, and voice standard produces misaligned feedback.
  10. Track recurring issues — If the same problems appear across multiple pieces, the issue is systemic. Flag it as a process improvement, not just an edit.

Integration Points

  • Copywriting — Use for writing new copy from scratch. Copy Editing handles reviewing and improving existing copy.
  • Content Humanizer — Use when AI-generated copy needs humanization before editorial review.
  • Content Production — Use Copy Editing as part of the production pipeline between drafting and publishing.
  • Brand Guidelines — Reference brand voice and style standards during the Voice and Tone sweep.
  • Marketing Psychology — Apply psychological principles during the Heightened Emotion sweep.
  • Content Strategy — Use when the problem is what to say, not how to say it.

Troubleshooting

ProblemLikely CauseFix
Same issues appear across multiple pieces from the same writerSystemic writing habit, not a one-off errorCreate a writer-specific checklist of recurring issues; address in style guide or training, not just per-piece edits
Copy loses its original voice after editingEditor over-corrected; replaced author voice with editor's stylePreserve author voice — enhance clarity and impact without rewriting personality. Read original aloud before editing
Edits introduce new inconsistenciesPrevious sweeps not re-checked after later sweeps modified contentAlways re-run earlier sweeps after making changes — Sweep 7 edits can break Sweep 1 clarity
CTA buried or ineffective despite multiple edit passesCTA was not the focus of any sweep — copy editing focused on prose qualityPrioritize CTA area first (Sweep 7: Zero Risk) before polishing earlier sections
Fact-checking reveals unverifiable claimsWriter invented statistics or used outdated dataFlag and return to writer — editor cannot invent evidence. Document all unverifiable claims explicitly
Style inconsistencies between sectionsMultiple writers contributed or copy was assembled from different draftsRun style consistency checklist end-to-end; standardize contractions, heading case, number format, and punctuation

Success Criteria

  • Seven Sweeps completion: All 7 sweeps completed per piece with previous sweeps re-verified after each pass
  • Error rate: Zero grammatical errors, typos, or broken links in published copy
  • Style consistency: 100% adherence to documented style guide (Oxford comma, heading case, number format, etc.)
  • Fact verification: All statistics have named source and year; all claims are verifiable or labeled as opinion
  • CTA effectiveness: Every piece has a clear, specific CTA visible within 2 scrolls of the content end
  • Clarity score: Every sentence understandable on first read — zero ambiguous pronoun references or multi-clause confusion
  • Edit turnaround: 48-hour maximum turnaround on editorial review with clear change documentation

Scope & Limitations

In scope:

  • Seven Sweeps editorial framework (clarity, voice, so-what, proof, specificity, emotion, zero-risk)
  • Quick-pass editing (word-level, sentence-level, paragraph-level)
  • Style consistency auditing and enforcement
  • Fact-checking protocol for claims, statistics, and competitive references
  • Pre-edit and post-edit checklists
  • Common copy problem diagnosis and fixing

Out of scope:

  • Writing new copy from scratch (use Copywriting skill)
  • AI content detection and humanization (use Content Humanizer)
  • SEO optimization passes (use Content Production optimization pipeline)
  • Content strategy or topic selection (use Content Strategy)
  • Visual design or layout feedback
  • Legal review of marketing claims

Known limitations:

  • Cannot verify internal company claims (product capabilities, uptime SLAs) without access to product documentation
  • Fact-checking external claims requires access to original sources — may need writer input
  • Style consistency requires an existing style guide; without one, editor must make judgment calls
  • Emotional impact (Sweep 6) is subjective and varies by audience — use target audience context
  • Multi-language copy editing requires native-level proficiency in each language

Scripts

# Score content readability with detailed metrics
python scripts/readability_scorer.py article.md --json

# Detect AI patterns that need humanization before editing
python scripts/ai_pattern_detector.py article.md --verbose

# Check style consistency across multiple documents
python scripts/style_checker.py --files doc1.md doc2.md doc3.md --json

平台分布

Codex

36.68%
按下载量换算204

Claude

30.25%
按下载量换算168

Cursor

19.83%
按下载量换算110

Gemini CLI

11.16%
按下载量换算62

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

通过

权限和风险

只读

该 Skill 主要提供规则、说明或参考内容,本身偏只读;真正读写文件、联网或执行命令仍取决于宿主 Agent 的任务。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。目前只有一个公开来源,建议同时核对源仓库。

继续浏览同类 Skills