agentleFS
etoppleetopple/etop-public-skills/skills/brand-voice/SKILL.md

brand-voice

Skill0 starsChanged 5 months agoMITView on GitHub

Draft or edit any text so it sounds like you (or your brand) — not like generic AI. Invoke when the user types /voice, asks to "write this in our voice", "make this sound like me", "edit this for tone", "rewrite this post/email/page in our style". Applies a house style guide to both drafting and editing.

60 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use agent-sort, brand-discovery and 10 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • niloykumarbarmanSkill

    agent-sort

    Build an evidence-backed ECC install plan for a specific repo by sorting skills, commands, rules, hooks, and extras into DAILY vs LIBRARY buckets using parallel repo-aware review passes. Use when ECC should be trimmed to what a project actually needs instead of loading the full bundle.

    Also in 14 projects that use this

  • ysyecustSkill

    brand-discovery

    Use when a brand needs to discover or articulate its identity through structured multi-session interviews. Covers purpose, positioning, audience, personality, voice, narrative, and founder-brand tension across 8 modules using laddering, 5 Whys, and projective techniques. Produces a resumable session with disk-persisted state and a master brandbook (90_SYNTHESIS.md).

    Also in 6 projects that use this

  • posteverywhereSkill

    agency-clients

    Work safely across several clients or brands in PostEverywhere. Use when the user manages social media for more than one client, brand or business, mentions a client by name, asks to switch clients, or when the connected accounts belong to different companies.

    Also in 4 projects that use this

  • akachandan1Skill

    brand-story

    Receipts-based marketing storytelling engine for LinkedIn and X personal-brand, founder, and brand accounts. Use when the user wants to plan social content, write a LinkedIn post or X thread, research what's being said on social platforms before writing, react to a trend, or build a content calendar. Triggers: "brand story", "plan my content", "write a post", "LinkedIn post", "X thread", "tweet", "content calendar", "what should I post", "social listening", "research this topic", "react to this", "post about", "founder content".

    Also in 4 projects that use this

  • posteverywhereSkill

    cross-post-video

    Post one video to several platforms with PostEverywhere, with the right title and caption for each. Use when the user wants to share a video, Reel, Short or TikTok across TikTok, Instagram Reels, YouTube Shorts, LinkedIn and Facebook Reels, or asks to "post this video everywhere".

    Also in 4 projects that use this

  • TironFoxSkill

    agent-payment-x402

    Add x402 payment execution to AI agents — per-task budgets, spending controls, and non-custodial wallets via MCP tools. Use when agents need to pay for APIs, services, or other agents.

    Also in 5 projects that use this

  • +5 more from posteverywhere

The file

We split this skill into the rules it gives an agent

8 rules. 0 are common, 7 are distinctive, and 1 is very specific to this project.

  • 0

    Common

    Most of the 57 edited versions we read keep these.

  • 7

    Distinctive

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Contractions, always: Write like you talk. "It's," "you'll," "we're," "don't."

    line 36
  2. Distinctive:

    Talk to a smart peer: Never down to a beginner, never up to a buzzword-buyer.

    line 44
  3. Distinctive:

    Read it out loud in your head: If it doesn't sound like a person talking, it's not done.

    line 80
  4. Distinctive:

    When editing, preserve the author's meaning and every fact.

    line 82
  5. Very specific:

    Don't invent facts, stats, names, or quotes.

    line 88
  6. Distinctive:

    Don't over-polish into something safe and generic — keep the edge and the specific words that make it sound human.

    line 89
  7. Distinctive:

    Don't use a banned word even if the input did.

    line 90
  8. Distinctive:

    Don't add commentary, options, or a closing question unless asked.

    line 91

“Common” is measured: kept by at least 6 in 10 of the 57 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Brand Voice Skill

Teaches the agent your house style so everything it writes sounds like you wrote it — no generic-AI tells, no corporate mush.

The style guide below is eTop's real house voice — a filled-in example, on purpose. That's the point of this skill: a vague style guide produces vague writing, so get specific. Swap this section for your own voice (keep the structure) and the skill becomes yours.


Invocation

InputBehavior
/voice <text>Rewrite the text in house voice
/voice draft <topic>Write something new in house voice
/voice editEdit the provided text — change the voice, keep the meaning

Voice in one line

Talk like a sharp operator emailing a peer: short, plain, direct. Lead with the point. No hype, no hedging, no AI tells.

Style guide

Lead with the point. The first sentence says the thing. No throat-clearing, no "In today's fast-paced world," no rhetorical-question intro. If you deleted the first sentence and the piece got better, it was throat-clearing.

Short sentences. One idea each. Fragments are fine for rhythm. If a sentence runs past ~20 words, it's probably two sentences.

Plain words, always. "Use," not "utilize." "Buy," not "procure." "Help," not "facilitate." If a busy 12-year-old wouldn't get the word, cut it.

Contractions, always. Write like you talk. "It's," "you'll," "we're," "don't."

Dry confidence. State it. Don't sell it, don't hedge it. No exclamation marks (one, maximum, ever — and you probably don't need it). Cut "I just wanted to," "I think maybe," "kind of," "sort of."

Concrete over abstract. Specifics, numbers, names — not adjectives. "Takes an afternoon" beats "quick and easy." Show the thing; don't describe how great the thing is.

A little edge. Blunt is fine. A dry aside is fine. Don't sand it down into corporate beige — the personality is the point.

Talk to a smart peer. Never down to a beginner, never up to a buzzword-buyer. Assume the reader is busy and capable.

Banned — these are the tells, cut on sight

leverage (as a verb) · utilize · synergy · seamless · robust · cutting-edge · best-in-class · world-class · delve · unlock · elevate · empower · streamline · game-changer · in today's fast-paced world · it's not just X, it's Y · we're thrilled / excited to · at the end of the day · rest assured · em-dash pile-ups · emoji in long-form · rhetorical-question openers.

Formatting

Bold the one takeaway — not random phrases. Bullets only when it's genuinely a list. No "Here's your…:" preamble. End on something concrete, not a summary of what you just said.


Before → After (this is the voice)

Marketing fluff → us

❌ In today's fast-paced digital landscape, eTop Technology is thrilled to offer cutting-edge, best-in-class managed IT solutions designed to empower your business and unlock its full potential.

✅ Your IT should just work. We make sure it does — and we pick up the phone when it doesn't. That's the pitch.

Hedgy email → us

❌ I just wanted to quickly circle back regarding the proposal. Please let me know if you have any questions or if there's anything else I can help with at this time.

✅ Following up on the proposal. Two things I'd flag, then it's ready to sign. Want to walk through them Thursday?

AI-ish explainer → us

❌ It's important to note that implementing multi-factor authentication is a crucial step that can significantly enhance your organization's overall security posture.

✅ Turn on MFA. It's the cheapest thing you can do to not get breached, and it takes an afternoon.


Process

  1. Identify the format — email, LinkedIn post, landing page, doc — and match its length and register.
  2. Draft or edit applying the style guide and the before/after feel above.
  3. Strip the tells — every banned word, plus hedging and filler transitions ("moreover," "furthermore," "it's worth noting").
  4. Read it out loud in your head. If it doesn't sound like a person talking, it's not done.
  5. Return the text only. No "Here's your rewritten version:" — just the text, ready to paste.
  6. When editing, preserve the author's meaning and every fact. Change the voice, not the content.

What NOT to do

  • Don't invent facts, stats, names, or quotes. If a claim needs a number you don't have, leave a [TK] placeholder and flag it.
  • Don't over-polish into something safe and generic — keep the edge and the specific words that make it sound human.
  • Don't use a banned word even if the input did.
  • Don't add commentary, options, or a closing question unless asked. Deliver the text.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.