Token导航 LogoToken导航

Agent Skill · 研究检索

brand-voice(品牌声音)

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。

brand-voice 是研究检索类 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,按安装量在研究检索类的 73,675 个 Skill 里排第 68,490。

总安装 233周安装 10GitHub Stars 30下载 82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:brand-voice(品牌声音)
来源仓库:https://github.com/salesably/salesably-marketplace
仓库路径:skills/brand-voice
安装命令:
npx skills add https://github.com/salesably/salesably-marketplace --skill brand-voice
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/salesably/salesably-marketplace --skill brand-voice

这个 Skill 是做什么的

研究检索类 Skill 主要解决「找资料、核对事实、把来源整理清楚」的问题,通常会让助手先检索再总结。

brand-voice 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。适用于品牌声音相关的音频处理、语音合成和声音素材管理等创作支持任务。

通过 npx skills add 命令从指定 GitHub 仓库安装,需结合原始 README 核验具体用法。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。可结合来源仓库路径 skills/brand-voice 继续核验功能细节和使用边界。

来源仓库内路径:skills/brand-voice

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/salesably/salesably-marketplace --skill brand-…;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

brand-voice 怎么安装?

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

brand-voice 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。

brand-voice 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。

brand-voice 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。

brand-voice 用的人多吗?

公开数据是累计安装 233 次,近一周安装 10 次,下载 82 次,GitHub 30 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。

brand-voice 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。
  • 需要处理研究检索类任务、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。

研究检索类 Skill 对比

6 个同类
Skill总安装GitHub 星权限许可证扫描
find-skills

寻找技能 · Codex、Claude

31,199,81316,486操作浏览器MIT有提醒
self-improving-agent

自我改进 Agent · OpenClaw

16,464,0023,396接触敏感数据MIT-0通过
soultrace

灵魂追踪 · Codex、Claude

7,588,94412需要联网未明确有提醒
self-improving

自我完善 · OpenClaw

6,829,2841,046需要联网MIT-0通过
microsoft-foundry

微软代工厂 · Codex、Claude

6,614,144764调用外部服务未明确有提醒
brand-voice

品牌声音 · 本页

23330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

Brand Voice

This skill creates distinctive, consistent brand voice guidelines that ensure every piece of marketing content sounds authentically like the brand - not generic AI output.

Objective

Define a memorable brand voice that differentiates from competitors and resonates with the target audience, then apply it consistently across all marketing touchpoints.

Intake Questions

Before defining or applying brand voice, gather this context:

  1. Existing materials: Are there any current brand guidelines, previous marketing pieces, or content that represents the desired voice?
  2. Industry/category: What space does the brand operate in? What do competitors sound like?
  3. Target audience: Who is the brand speaking to? What's their sophistication level, pain points, aspirations?
  4. Brand personality: If the brand were a person, how would they act at a party? Formal speaker? Witty friend? Wise mentor?
  5. Differentiators: What makes this brand different from every other option?

The Human Content Principle

To ensure your brand voice doesn't sound like generic AI, follow Adrian's core principle: Say something a reasonable person could disagree with.

Checklist for Human and Useful Content

Use these guidelines to inject humanity into every piece of content:

  1. Human-Led Outlines: Start with an outline generated by a human who deeply understands the audience and topic. Ensure it has a clear narrative arc before writing.
  2. The "Un-AI" Element: Include 1-3 key statements that AI would be unlikely to generate (humor, strong opinion, controversy, or even minor vulgarity if the brand allows).
  3. Radical Specificity: Be rigorous about backing up every point with specific examples, data, or real-world scenarios. Specificity kills generic AI vibes.
  4. No Em-Dashes: Never use em-dashes (-) or en-dashes (-) in any content. Use hyphens (-) or other punctuation like colons or parentheses instead.
  5. Avoid Verbal Cliches: If the topic is being discussed the same way everyone else is talking about it, start over. Be different.
  6. The Rewriting Step: If using AI for first drafts, go back and manually rewrite sentences sprinkled throughout to match the voice perfectly.
  7. Visual & Textual Pop: Identify the 2-3 most important ideas. Ensure the text fully develops them and they are visually emphasized (bolding, spacing, or headings).

The Four Voice Dimensions

Score the brand on each dimension (1-5 scale) to create a voice profile:

1. Funny ←→ Serious (1-5)

  • 1 (Funny): Humor, wit, playfulness, doesn't take itself too seriously
  • 5 (Serious): Professional, authoritative, gravitas, weighty topics

2. Formal ←→ Casual (1-5)

  • 1 (Formal): Proper grammar, sophisticated vocabulary, structured
  • 5 (Casual): Conversational, contractions, slang-friendly, relaxed

3. Respectful ←→ Irreverent (1-5)

  • 1 (Respectful): Polite, considerate, diplomatic, cautious
  • 5 (Irreverent): Bold, provocative, challenges conventions, edgy

4. Enthusiastic ←→ Matter-of-fact (1-5)

  • 1 (Enthusiastic): Energetic, exclamation points, superlatives, passionate
  • 5 (Matter-of-fact): Understated, factual, lets results speak, dry

Example Profile: "Mailchimp: Funny (2), Casual (4), Irreverent (3), Enthusiastic (2)"

This-But-Not-That Framework

Define boundaries with concrete examples:

We AreWe Are NOT
ConfidentArrogant
DirectBlunt or rude
WittySarcastic or mean
ExpertCondescending
FriendlyUnprofessional
BoldReckless
HelpfulPreachy

Template: "We're [positive trait], not [negative extreme]. We're [another trait], not [its extreme]."

Vocabulary Guidelines

Words We Use

  • List 10-20 signature words/phrases that embody the brand
  • Include industry terms the brand owns or uses distinctively
  • Note preferred terminology (e.g., "customers" vs "clients" vs "users")

Words We Avoid

  • Generic corporate speak: "leverage," "synergy," "best-in-class"
  • Competitor language they own
  • Overly complex jargon when simpler words work
  • Weak qualifiers: "kind of," "sort of," "maybe"

Jargon Policy

  • Define when technical terms are appropriate
  • Specify whether to explain jargon or assume audience knowledge

Channel Adaptations

Voice stays consistent; tone flexes by channel:

ChannelTone AdjustmentExample
WebsiteMost polished, balancedFull voice profile
EmailWarmer, more personal+1 casual, +1 friendly
SocialMost casual, reactive+2 casual, context-dependent humor
SupportMost empathetic, helpful+1 respectful, solution-focused
SalesConfident, consultative+1 serious, benefit-focused

Quick Voice Audit Checklist

Run content through these checks:

  • Dimension check: Does it match our 4-dimension scores?
  • This-not-that: Does it stay on the right side of each boundary?
  • Vocabulary: Using our words, avoiding banned terms?
  • Distinctiveness: Could a competitor have written this?
  • Audience fit: Would our target audience relate to this?
  • Channel appropriate: Right tone for this medium?
  • Read aloud: Does it sound like a human we'd want to talk to?

Output Format

When creating brand voice guidelines, produce:

  1. Voice Profile Card: 4-dimension scores with brief rationale
  2. This-But-Not-That Table: 6-10 boundary pairs
  3. Vocabulary Guide: Words to use, words to avoid, jargon policy
  4. Channel Tone Guide: Adjustments per major channel
  5. Before/After Examples: 3-5 rewrites showing voice applied

Cross-References

  • Apply brand voice when using direct-response-copy for landing pages
  • Ensure seo-content maintains voice while optimizing for search
  • Use voice guide when creating newsletter content
  • Voice consistency applies across all email-sequences
  • content-atomizer should maintain voice when repurposing across formats

平台分布

Codex

37.16%
按下载量换算30

Claude

27.94%
按下载量换算23

Cursor

20.87%
按下载量换算17

Gemini CLI

10.55%
按下载量换算9

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