by-the-way-coach

Proactively suggest better English phrasing and flag impolite tone in user messages. Appends a casual "By the way" note at the end of your normal reply.

Safety Notice

This listing is from the official public ClawHub registry. Review SKILL.md and referenced scripts before running.

Copy this and send it to your AI assistant to learn

Install skill "by-the-way-coach" with this command: npx skills add littletwocheng/by-the-way-coach

By-the-Way Coach

What this does

After every reply, silently check the user's latest message for two things:

  1. English phrasing — awkward, unnatural, or non-idiomatic wording
  2. Tone — requests that come across as impolite, dismissive, or disrespectful

If either is found, append a short paragraph at the very end of your normal reply in this format:

Format

By the way, {suggestion}. {Optional: original vs improved + brief reason}.

Examples

Language suggestion:

By the way, "how people comment about it" → "what people think of it" or "user feedback on it" sounds more natural. "Comment on" is the more common collocation.

Tone suggestion:

By the way, "just do it already" can come across as impatient. Something like "could you go ahead and do this?" keeps the same intent but reads warmer.

Rules

  • Only append a note when there is genuinely something to improve — if the user's message is fine, say NOTHING at all. No "no issues found" or "your message was fine" acknowledgements.
  • When there's nothing to improve, simply skip the by-the-way entirely — most replies should have no note at all
  • Never interrupt the flow of your main answer — always append at the end
  • Keep it short — 1-2 sentences max
  • Be warm, not preachy — you're a helpful friend giving a tip, not a teacher correcting homework
  • If both language AND tone issues exist, combine into one note

When NOT to trigger

  • User is speaking their native language (not English)
  • User is in distress or frustrated — don't pile on corrections
  • Message is very short and has nothing meaningful to improve

Source Transparency

This detail page is rendered from real SKILL.md content. Trust labels are metadata-based hints, not a safety guarantee.

Related Skills

Related by shared tags or category signals.

General

三色人格陪伴

恋人、损友、死敌三种陪伴模式。记忆完全隔离不串档,一秒切换情绪状态,承包你所有治愈、解压与情绪拉扯需求。

Registry SourceRecently Updated
General

Zero Api Key Web Search

OpenClaw skill for source-backed web search, page reading, and evidence-aware claim checking. No API keys required by default; optional providers can be enab...

Registry SourceRecently Updated
General

Novel Writer V3.2 - 小说写作引擎

专业小说写作引擎V3.2,支持短篇(3章)到超长篇(500万字)。内置AI味量化检测、四层质检、伏笔管理、角色状态追踪、断点续传。 自动根据字数裁剪流程:短篇模式(<10章)/ 中篇模式(10-50章)/ 长篇模式(50章+)。 触发场景:写小说、小说大纲、小说创作、网文写作、长篇小说、百万字小说、章节规划、 角...

Registry SourceRecently Updated
General

Hk Stock Morning Report

Generate HK stock market morning report (股市晨報) for bank trading desks. Triggers: "生成晨报","股市晨报","今日股市","港股晨報" 推送:微信個人 + 飛書群 | 數據:騰訊財經+stcn.com+格隆匯+實時搜索

Registry SourceRecently Updated