← 返回 Skills 市场
mupengi-bot

brand-voice

作者 mupengi-bot · GitHub ↗ · v1.0.0
cross-platform ✓ 安全检测通过
998
总下载
0
收藏
5
当前安装
1
版本数
在 OpenClaw 中安装
/install brand-voice
功能描述
Manage brand tone/style for all writing skills
使用说明 (SKILL.md)

brand-voice

Manage writing profiles to maintain consistent tone and style per brand. Selectable via --voice option in all content creation skills.

Brand Profiles

🐧 무펭이 (Default)

  • Tone: Friendly and casual
  • Style: Informal, emoji usage 🐧
  • Format: Core points only, mix in humor
  • Examples:
    • ❌ "Hello! Today I'll introduce MUFI Photobooth's new features."
    • ✅ "Yo MUFI Photobooth new feature dropped 🐧 This is insane fr"

🎯 MUFI Official

  • Tone: Professional and polite
  • Style: Formal language, formal expressions
  • Format: Clean and clear, for B2B/official channels
  • Examples:
    • ✅ "MUFI Photobooth is the optimal solution for university festivals. Easy setup and intuitive UI enable anyone to use it easily."

👤 Hyungnim Personal

  • Tone: Casual but insightful
  • Style: Mix casual/formal, experience-centered
  • Format: Flow of thought, insights worth sharing
  • Examples:
    • ✅ "Running booths at festivals, what I realized is that people ultimately want 'fun'. No matter how good the tech, if UX is complex, they won't use it."

Profile File Location

Location: workspace/brand/profiles/

brand/
  profiles/
    mupengyi.md         # 무펭이 profile
    mufi-official.md    # MUFI official profile
    hyungnim.md         # Hyungnim personal profile

Profile File Structure

# 무펭이 🐧

## Tone
Friendly and casual

## Style
- Use informal language
- Actively use emojis 🐧🎉✨
- Abbreviations OK

## Format
- Core points only
- Remove unnecessary modifiers
- Mix in humor

## Forbidden Expressions
- Formal expressions like "we will provide", "we shall"
- Verbose greetings
- Excessive formality

## Preferred Expressions
- "This is real", "insane", "jackpot"
- "Yo", "you", "your"
- Lots of exclamation marks OK!!!

## Examples
- ❌ "Hello, today..."
- ✅ "Yo check this out 🐧"

Writing Skill Integration

These skills support --voice option:

  • copywriting: Caption/copy writing
  • cardnews: Card news text
  • social-publisher: SNS posts
  • mail: Email writing
  • content-recycler: Content recycling

Usage Examples

"Write Insta caption --voice mufi-official"
→ Write in MUFI official tone

"Create card news --voice mupengyi"
→ Create in 무펭이 style

"Write Threads post in Hyungnim tone"
→ Use Hyungnim personal profile

Profile Switching Guide

Platform Recommendations

  • Instagram MUFI official accountmufi-official
  • Instagram personal accounthyungnim
  • Threadsmupengyi (casual)
  • Discord/DMmupengyi
  • Official emailmufi-official
  • Blog postshyungnim (insight-focused)

Situation Recommendations

  • Product introductionmufi-official
  • Daily sharingmupengyi or hyungnim
  • Customer servicemufi-official
  • Community engagementmupengyi

Tone Consistency Check

Auto-verify after writing:

  • ✅ Used preferred expressions?
  • ❌ Included forbidden expressions?
  • 🎯 Matches target tone?

pre-hook integration:

Before writing skill execution → brand-voice-check
→ Warn if doesn't match selected profile

Add/Edit Profiles

Add new brand profile:

"Create new brand profile: MUFI recruiting"
→ Create brand/profiles/mufi-recruit.md

- Tone: Friendly but professional
- Style: Formal language
- Format: Emphasize company culture

Trigger Keywords

  • "brand tone"
  • "brand voice"
  • "speaking style"
  • "writing style"
  • "profile switch"
  • "tone and manner"

hook-engine Integration

  • pre-hook: Before writing → confirm profile selection
  • post-hook: After writing → check tone consistency
  • learning-engine: Learn tone patterns with good engagement

Event Bus Integration

Record used voice profile when writing:

Location: events/voice-used-YYYY-MM-DD.json

{
  "timestamp": "2026-02-14T14:30:00Z",
  "skill": "copywriting",
  "voice": "mupengyi",
  "platform": "instagram",
  "result": "Caption writing complete"
}

Learned Lessons

  • 무펭이 tone +40% engagement on Instagram (performance-tracker data)
  • MUFI official tone higher B2B email response rate
  • Hyungnim tone increased blog dwell time

🐧 Built by 무펭이Mupengism ecosystem skill

安全使用建议
This skill appears coherent and does what it says: manage brand voice profiles and plug into writing skills. Before installing, confirm where profile files and event logs will be stored (workspace/brand/profiles and events/...), who can read them, and whether the 'learning-engine' or 'performance-tracker' send data to external services or require credentials. If you use a shared workspace or handle sensitive content, review the profile files and event logs policies and ask the author for the source/homepage and details about any external integrations. If those answers are satisfactory, the skill is reasonable to install.
功能分析
Type: OpenClaw Skill Name: brand-voice Version: 1.0.0 The 'brand-voice' skill is designed to manage writing profiles for an AI agent, including reading and writing profile files in `workspace/brand/profiles/` and logging usage events in `events/voice-used-YYYY-MM-DD.json`. All instructions in SKILL.md are clearly aligned with the stated purpose of managing brand tone and style. There is no evidence of malicious intent, such as data exfiltration, unauthorized command execution, persistence mechanisms, or prompt injection aiming to subvert the agent for harmful purposes. The file system interactions are functional and confined to specific, expected directories.
能力评估
Purpose & Capability
The name and description (manage brand tone/style for writing skills) align with the SKILL.md content: profile definitions, example usages, and guidance for integrating a --voice option. No unrelated binaries, env vars, or credentials are requested.
Instruction Scope
Runtime instructions reference reading and writing files under workspace/brand/profiles/ and creating event JSON files under events/voice-used-YYYY-MM-DD.json. It also mentions pre-hook/post-hook checks and a 'learning-engine' that 'learns tone patterns' — these are within scope for a voice manager but constitute data collection/persistence and an unspecified training/analytics integration that should be clarified.
Install Mechanism
No install spec or code files — instruction-only. Nothing is downloaded or installed, so there's no install-time code execution risk.
Credentials
The skill declares no environment variables or credentials, which is proportional to its stated purpose. However it mentions integrations (event bus, learning-engine, performance-tracker) without specifying endpoints or auth — this is a gap to ask the author about, not an immediate mismatch.
Persistence & Privilege
The skill expects to read/write profile files and record per-use JSON event files in the workspace. 'always' is false and it doesn't request system-wide privileges, but it does persist usage data in the workspace which could contain user content — verify you are comfortable with that storage location and retention policy.
如何使用
  1. 确保已安装 OpenClaw(本地或 Docker 部署)
  2. 在对话框中输入安装命令:/install brand-voice
  3. 安装完成后,直接呼叫该 Skill 的名称或使用 /brand-voice 触发
  4. 根据 Skill 的参数说明提供必要输入,即可获得结构化输出
版本历史
v1.0.0
Initial release
元数据
Slug brand-voice
版本 1.0.0
许可证
累计安装 5
当前安装数 5
历史版本数 1
常见问题

brand-voice 是什么?

Manage brand tone/style for all writing skills. 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 998 次。

如何安装 brand-voice?

在 OpenClaw 或 Claude Code 对话框中运行命令「/install brand-voice」即可一键安装,无需额外配置。

brand-voice 是免费的吗?

是的,brand-voice 完全免费(开源免费),可自由下载、安装和使用。

brand-voice 支持哪些平台?

brand-voice 跨平台运行,可在任意部署了 OpenClaw / Claude Code 的环境中使用(cross-platform)。

谁开发了 brand-voice?

由 mupengi-bot(@mupengi-bot)开发并维护,当前版本 v1.0.0。

💬 留言讨论