Token导航 LogoToken导航TokenDH.com
研究检索只读github未标认证来源可访问许可证需确认审计异常

voice-editor语音编辑器

Agent Skill

用于辅助音频、音乐、语音转写、语音合成或声音素材处理。它适合让 Agent 生成配乐说明、整理音频流程、调用语音工具或处理播客和视频配音素材。使用时需要确认输入音频来源、输出格式、时长和模型限制;涉及人声克隆、版权音乐或公开发布时,应先核对授权和合规边界。

总安装

256

周安装

11

GitHub Stars

1

下载量

90
CodexClaudeCursorGemini CLI

安装说明

本站只整理中文说明和来源信息,不托管安装包,也不代用户安装。

GitHub

来源数

2

许可证

unknown

最后核验

2026-05-01

来源状态

来源可访问

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:voice-editor(语音编辑器)
来源仓库:https://github.com/aplaceforallmystuff/claude-voice-editor
仓库路径:skills/voice-editor
安装命令:
npx skills add https://github.com/aplaceforallmystuff/claude-voice-editor --skill voice-editor
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/aplaceforallmystuff/claude-voice-editor --skill voice-editor

简介

用于辅助音频、音乐、语音转写、语音合成或声音素材处理,适合生成配乐说明或整理音频流程。

  • 适用于调用语音工具、处理播客和视频配音素材等场景。
  • 通过 npx skills add 命令从指定 GitHub 仓库安装并使用。
  • 使用时需确认输入音频来源、输出格式及时长限制,注意版权合规边界。
  • voice-editor 属于研究检索类 Skill,可作为该场景下的辅助能力补充。

SKILL.md

Voice doesn't live in first drafts -- it lives in editing choices. This skill guides the 6-pass editing process that turns AI's generic output into something distinctively yours.

<quick_start> Provide:

  1. Content to edit (paste or file path)
  2. VOICE.md location (default: ~/.claude/voice/VOICE.md or project's .claude/VOICE.md)

The skill will:

  1. Run slop detection on the content
  2. Identify voice mismatches against your profile
  3. Perform multi-pass editing (Structure → Voice → Polish)
  4. Output edited content with change annotations
  5. Suggest VOICE.md updates based on patterns found </quick_start>

<the_editing_reality> The 30-40% Rule:

Even with a perfect style guide, expect to edit 30-40% of AI output to make it distinctively yours. This isn't a problem -- it's the feature.

30-40% editing effort is dramatically less than 100% writing-from-scratch effort, while producing work that sounds more like you than unguided AI ever could.

If you're editing more than 50%: Your style guide needs refinement, or your prompts need work.

If you're editing less than 20%: You might be accepting too much generic content. </the_editing_reality>

<editing_workflow> Focus: Does the organization serve the content?

Opening Mechanics (from craft-rules):

  • Does opening create a SPECIFIC question? Not vague mystery - readers need enough context to form a particular question they want answered
  • Does first paragraph contain change or threat of change? Brains notice disruption of status quo
  • Is first paragraph under 50 words / 3 sentences?
  • First line hits 3+ of: question, character, setting, tone, movement?

Questions to ask:

  • Does the opening hook or just clear throat?
  • Is there a clear progression of ideas?
  • Does each section earn its place?
  • Is anything redundant or missing?
  • Does the conclusion land or just stop?

Scene/Section Structure:

  • Does each section follow Goal → Conflict → Outcome?
  • Are transitions connected by "because" not "and then"?
  • Are characters/subjects in motion during slow passages?

Actions:

  • Cut throat-clearing openings ("In this article, we'll explore...")
  • Reorder sections for better flow
  • Merge redundant sections
  • Add missing transitions
  • Strengthen weak conclusions
  • Add change to opening if missing - start with what changed, not background
  • Make question specific - "Why is the other side of the bed cold?" not "Something is happening"

Check against VOICE.md:

  • Sentence length patterns (matching variance?)
  • Vocabulary fingerprint (right formality level?)
  • Tone markers (humor, directness, reader relationship?)
  • Structural habits (lists vs prose, formatting?)
  • Opinion expression (appropriate confidence level?)

Actions:

  • Replace formal language with conversational equivalents
  • Vary sentence lengths to match profile
  • Add characteristic phrases from the voice profile
  • Adjust opinion strength to match profile
  • Fix paragraph rhythm

Scan for:

  • Tier 1 phrases (remove immediately)
  • Tier 2 patterns (check context)
  • Tier 3 clusters (break up if grouped)
  • Structural tells (uniformity, over-balance)
  • Staccato fragment spam (3+ consecutive short declarative sentences)
  • Comparator sentences ("This isn't X. It's Y." patterns)

Actions:

  • Delete or replace all Tier 1 phrases
  • Evaluate Tier 2 in context, remove if generic
  • Redistribute Tier 3 to avoid clustering
  • Vary sentence lengths if too uniform
  • Rebalance sections if all equal weight
  • Combine staccato fragments into flowing prose with commas, em-dashes, or conjunctions
  • Rewrite comparators directly -- just say what something IS, don't waste words on what it isn't

What to cut:

  • Generic superlatives without claims ("game-changing," "revolutionary")
  • Hedging that adds no value ("it might be argued that")
  • Corporate platitudes ("leverage synergies")
  • Anything that sounds like a press release

What to add:

  • Your actual opinion (not the diplomatic version)
  • Specific examples from your experience
  • Honest acknowledgment of limitations
  • Your characteristic quirks and phrases
  • Conversational asides that show how you think

The read-aloud test:

  • Read the piece aloud (or use text-to-speech)
  • Mark every place you stumble
  • Note where you'd naturally pause vs. where punctuation forces pauses
  • Identify sentences that feel too long or choppy

Sentence Craft Checks (from craft-rules):

  • Active verbs predominate? Search for "was [verb]ed" passive patterns
  • Word order follows action sequence? Actor → Action → Recipient (filmic)
  • Objects have ~3 specific qualities, not abstract adjectives?
  • Sentence length varies deliberately? Short punch, then longer exploration
  • No clichéd metaphors? (Fresh comparisons trigger more brain activity)

Hemingway Principles:

  • Emphasize nouns
  • Choose active verbs
  • Adjectives: few but apt
  • Short sentences by default
  • Long sentences ONLY for: speeding action, describing flow, making short sentences pop

Actions:

  • Break sentences where you ran out of breath
  • Combine choppy sentences that flow together
  • Add rhythm variation (short punch, longer exploration)
  • Fix awkward transitions
  • Smooth unnatural phrasing
  • Convert passive to active - "The ball was thrown by John" → "John threw the ball"
  • Reorder for transitive flow - "Jane gave a kitten to her Dad" beats "Jane gave her Dad a kitten"
  • Replace abstract with specific - "terrible" → details that create terror

Ask yourself: "Would someone who knows me recognize this as mine before seeing my name?"

If no:

  • What's missing? Add it.
  • What's wrong? Fix it.
  • What's generic? Make it specific.

If yes:

  • You're done. Ship it.

</editing_workflow>

<what_to_cut> Always remove:

  • Generic superlatives without specific claims
  • Hedging language that softens everything
  • Unnecessary qualification that adds no value
  • Corporate platitudes that mean nothing
  • Anything that sounds like a committee wrote it
  • Motivational poster language
  • Flattery sandwiches ("While X has merit, Y offers...")

Examples:

  • "This groundbreaking approach revolutionizes..."
  • "This cuts processing time from 4 hours to 15 minutes"
  • "It's worth noting that in many cases..."
  • [Just make the point directly]
  • "Unlock your potential with powerful insights"
  • "Stop wasting time on tasks AI handles in seconds" </what_to_cut>

<what_to_add> Voice elements AI misses:

  • Your actual opinion - Not the diplomatic version AI produces
  • Specific examples - From your experience, not generic training data
  • Honest limitations - Not fake confidence in uncertain claims
  • Characteristic quirks - Phrases your style guide might have missed
  • Conversational asides - How you actually think through problems
  • Rough edges - The imperfections that make writing human

Examples:

  • Add: "Three hours later (no hyperbole), I finally got it working"
  • Add: "Here's where I screwed up the first time..."
  • Add: "Look, I'm not sure this is right, but..."
  • Add: [Your characteristic phrases from VOICE.md] </what_to_add>

<output_format> For edited content, provide:

## Edited Content

[Full edited content here]

---

## Edit Summary

**Passes completed:** 6/6
**Estimated edit percentage:** [X]%
**Voice match confidence:** [Low/Medium/High]

### Major Changes
1. [Most significant change and why]
2. [Second major change]
3. [Third major change]

### Slop Removed
- [List of AI patterns found and removed]

### Voice Elements Added
- [List of voice-specific additions]

### Remaining Concerns
- [Any areas that still feel off]
- [Suggestions for further refinement]

### VOICE.md Update Suggestions
- [Patterns discovered that should be added to voice profile]
- [Phrases that worked well - consider adding to "use freely" list]

</output_format>

<editing_modes> Use when: Content is close to voice, just needs polish

Passes: 3, 5, 6 (Slop Removal, Rhythm, Final Check) Expected edit: 10-20% Time: 5-10 minutes per 500 words

Passes: All 6 Expected edit: 30-40% Time: 15-20 minutes per 500 words

Passes: All 6, possibly multiple iterations Expected edit: 50-70% Time: 25-35 minutes per 500 words

Consider: Might be faster to rewrite with better prompts

Approach:

  1. Extract the core ideas/structure only
  2. Rewrite from scratch using voice profile
  3. Incorporate any specific facts/examples from original
  4. Run standard editing passes on new draft

This is not editing -- it's salvage. </editing_modes>

<common_problems> Symptom: Reads like a corporate memo Cause: AI defaulting to academic/business training data Fix:

  • Replace "individuals" → "people"
  • Replace "utilize" → "use"
  • Replace "implement" → "do/start"
  • Add contractions
  • Shorten sentences
  • Add conversational asides

<integration_with_other_skills> Before voice-editor:

  • Run slop-detector for initial assessment
  • Ensure voice-analyzer has created VOICE.md
  • Run craft-rules for structural quality audit

After voice-editor:

  • Run slop-detector again to verify cleanup
  • Update VOICE.md with new patterns discovered

Full Workflow:

[Draft] → slop-detector → craft-rules → voice-editor → slop-detector → [Final]
               ↓               ↓              ↓               ↓
         AI tells?      Structure ok?   Voice match      Verify clean

What each skill checks:

SkillFocusExample Issue
slop-detectorAI patterns"delve," "landscape," "game-changer"
craft-rulesWriting mechanicsWeak hook, passive voice, abstract adjectives
voice-editorPersonal voiceToo formal, missing characteristic phrases

Quick workflow (already clean content):

[Draft] → voice-editor (light mode) → [Final]

Heavy workflow (generic AI output):

[Draft] → slop-detector → craft-rules → voice-editor (heavy/rescue) → slop-detector → [Final]

</integration_with_other_skills>

<success_criteria> Editing is complete when:

  • All 6 passes have been applied (or appropriate subset for mode)
  • No Tier 1 slop phrases remain
  • Sentence length varies naturally
  • Content matches VOICE.md characteristics
  • Read-aloud test passes (no stumbling)
  • The authenticity question is answered "yes"
  • Edit summary documents all changes
  • VOICE.md update suggestions provided if patterns discovered

The ultimate test: Would someone who knows the author recognize this as theirs before seeing the name? </success_criteria>

<core_principle> The style guide gives AI direction -- it helps produce output closer to your voice than generic defaults. Your editing gives the work your actual voice -- the quirks, opinions, personality, and rough edges that make it distinctively yours.

This is why "zero effort" AI fails. The promise of finished content without touching it misunderstands what makes writing valuable. Effort isn't the enemy. Wasted effort is.

You're redirecting effort from blank-page generation to voice-preservation editing. That redirection is the entire point. </core_principle>

适合场景

01

用户想查找某类 Agent Skill 时

02

需要根据任务场景推荐可安装能力包时

03

需要对比不同来源的安装命令和来源信息时

能力概览

能力 1

按任务关键词查找相关 Skills

能力 2

展示可复制的安装命令

能力 3

保留来源站点、仓库和原始说明,方便继续核验

能力 4

展示第三方安全扫描或审计结果

安装后应在对应宿主中按原始 README 的触发条件使用;具体调用方式请以来源页面和 README 为准。

平台分布

Codex

39.52%
按下载量换算36

Claude

27.26%
按下载量换算25

Cursor

17.75%
按下载量换算16

Gemini CLI

9.57%
按下载量换算9

安全审计

Gen Agent Trust Hub

通过

Socket

未通过

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。来源安全扫描存在 warning/failed 结果,不能写成本站确认安全。当前只有一个来源,正式发布前建议补源仓库或其他目录站核验。

来源信息

继续浏览同类 Skills