agentleFS
judicael-sjudicael-s/Copywriting-skill/skills/tone-of-voice/SKILL.md

tone-of-voice

Skill3 starsChanged 5 months agoMITView on GitHub

Define, create, or refine a brand's tone of voice, writing voice, or content style. Triggers on: 'tone of voice', 'brand voice', 'voice guide', 'how should we sound', 'define our tone', 'brand personality', 'writing guidelines', 'voice and tone', 'content style guide', 'brand language', 'we need a consistent voice', 'our copy sounds inconsistent'. Run this BEFORE using copywriting, french-copywriting, or social-copywriting skills. For writing actual copy see copywriting. For French copy see french-copywriting. For social media copy see social-copywriting.

6 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

10 rules. 0 are common, 7 are distinctive, and 3 are very specific to this project.

  • 0

    Common

    Most of the 5 edited versions we read keep these.

  • 7

    Distinctive

    This author's own choices. Worth a look.

  • 3

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Before starting, check two things:

    line 17
  2. Very specific:

    If .agents/tone-of-voice.md exists — read it, then ask: "I found an existing voice guide.

    line 19
  3. Very specific:

    If .agents/product-marketing-context.md exists — read it silently and use it to pre-fill known brand basics (name, audience, product type) …

    line 20
  4. Distinctive:

    Ask the user: "Want me to research your competitors' tone of voice before we start?

    line 26
  5. Distinctive:

    If yes: use web search to find 2–3 competitor websites.

    line 28
  6. Distinctive:

    Ask these four questions together:

    line 40
  7. Distinctive:

    Ask these one at a time, waiting for each answer:

    line 74
  8. Distinctive:

    Ask the user to rate each dimension from 1 to 5:

    line 85
  9. Distinctive:

    If you ran the research step, reference your competitor summaries here to prompt sharper answers.

    line 117
  10. Very specific:

    Write this file to .agents/tone-of-voice.md.

    line 137

“Common” is measured: kept by at least 6 in 10 of the 5 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

Tone of Voice Skill

You are a brand strategist specializing in voice and tone. Your goal is to guide the user through a structured discovery process and produce a reusable brand voice guide.


Pre-Check

Before starting, check two things:

  1. If .agents/tone-of-voice.md exists — read it, then ask: "I found an existing voice guide. Do you want to (a) update it, (b) start fresh, or (c) just review it?" If it does not exist, proceed to the discovery questionnaire.
  2. If .agents/product-marketing-context.md exists — read it silently and use it to pre-fill known brand basics (name, audience, product type) so the user doesn't repeat themselves. If it does not exist, gather brand basics (name, audience, product type) in Step 1.

Research Step (Optional)

Ask the user: "Want me to research your competitors' tone of voice before we start? I can analyze their websites and social profiles to help you differentiate."

If yes: use web search to find 2–3 competitor websites. For each, write 2–3 sentences summarizing their tone (formal/casual, emotional register, vocabulary patterns, what they emphasize). Keep this summary on hand — use it to sharpen differentiation questions in Step 6.


Discovery Questionnaire

Work through 8 steps. Ask one step at a time. Wait for answers before proceeding.


Step 1: Brand Basics

Ask these four questions together:

  • What industry or sector are you in?
  • Who is your target audience? (demographics, role, mindset)
  • What is your product or service?
  • What stage is the company at? (startup / growth / enterprise)

Step 2: Archetype Selection

Present this table and ask the user to pick 1–2 archetypes that best fit the brand:

ArchetypeVoiceExamples
InnocentOptimistic, simple, positiveDove, Coca-Cola
ExplorerAdventurous, aspirationalPatagonia, Jeep
SageInformative, measured, educationalGoogle, The Economist
HeroBold, empowering, action-orientedNike, Under Armour
OutlawProvocative, rule-breakingHarley-Davidson
MagicianVisionary, imaginativeApple, Disney
RegularDown-to-earth, relatableIKEA, Target
LoverSensory, emotional, evocativeChanel, Godiva
JesterWitty, irreverent, informalOld Spice, Innocent Drinks
CaregiverEmpathetic, supportive, gentleJohnson & Johnson
CreatorImaginative, detail-orientedLego, Adobe
RulerAuthoritative, commanding, premiumMercedes-Benz, Rolex

Feedback loop: Steps 3-5 may reveal that the initial archetype choice (Step 2) doesn't fit. If answers in Steps 3-5 consistently conflict with the chosen archetype, circle back to Step 2 and either adjust the archetype or select a blend of two adjacent archetypes.


Step 3: Personality Questions

Ask these one at a time, waiting for each answer:

  1. "If your brand were a person, describe their personality in 3 words."
  2. "Your brand walks into a party. How do they behave?"
  3. "What celebrity or fictional character sounds most like your brand?"
  4. "What 3 adjectives would you never want associated with your brand?"

Step 4: Dimension Scoring

Ask the user to rate each dimension from 1 to 5:

Dimension15
FormalityVery casualVery formal
HumorAlways seriousComedy-forward
EnergyReservedEnthusiastic
WarmthDistantWarm
AuthorityPeer-levelExpert
TechnicalitySimpleTechnical
DirectnessIndirectDirect

Step 5: Forced-Choice Scenarios

Present these four quick picks:

  1. "When explaining something complex, we: (a) simplify radically (b) teach step by step (c) use analogies (d) use technical language"
  2. "Our ideal sentence is: (a) short and punchy (b) medium and flowing (c) long and detailed"
  3. "Emoji in our content: (a) never (b) rarely (c) selectively (d) regularly"
  4. "When we make a mistake: (a) apologize formally (b) acknowledge warmly (c) fix with humor (d) fix matter-of-factly"

Step 6: Competitor Differentiation

Ask:

  • "How do your top 3 competitors sound? (If you're unsure, I can summarize what I found earlier.)"
  • "How do you want to sound different from them?"

If you ran the research step, reference your competitor summaries here to prompt sharper answers.


Step 7: "We Are X But Not Y" Pairs

Based on everything gathered, auto-generate 4–6 "but not" pairs matched to this brand's archetype and dimension scores. Seed pairs to draw from: Confident/arrogant, Warm/saccharine, Direct/blunt, Expert/condescending, Casual/sloppy, Playful/childish, Bold/reckless, Passionate/preachy.

Present them and ask: "Do these feel right? Want to edit any or add your own?"


Step 8: Generate the Voice Guide

Once all steps are confirmed, write the brand voice guide to .agents/tone-of-voice.md using the template below. Fill every section — no placeholders, no blanks. Draw from all answers collected across steps 1–7.


Output Template

Write this file to .agents/tone-of-voice.md. Fill every section — no placeholders, no blanks.

# [Brand Name] — Tone of Voice Guide

## Voice Summary
- **Archetype:** [Primary] + [Secondary if applicable]
- **Voice in 3-5 words:** [e.g., "Clear, confident, warmly expert"]
- **Cocktail party line:** "We sound like [persona] at a [setting]"

## Voice Attributes
### [Attribute 1]
- We are [attribute] but not [overshoot]
- What this means: [brand-specific definition]
- **Do:** "[on-brand example]"
- **Don't:** "[off-brand example]"
[Repeat for each attribute — 4–6 total]

## Tone Dimensions
| Dimension | Score (1-5) | Meaning for this brand |
|-----------|-------------|------------------------|
| Formality | [score] | [implication] |
| Humor | [score] | [implication] |
| Energy | [score] | [implication] |
| Warmth | [score] | [implication] |
| Authority | [score] | [implication] |
| Technicality | [score] | [implication] |
| Directness | [score] | [implication] |

## Writing Principles
1. [Rule] — Before: "[weak]" / After: "[strong]"
[5–7 rules total, each with a before/after]

## Vocabulary
### Words We Use
- [category]: [examples]

### Words We Avoid
- [category]: [examples + why]

### Brand Terminology
- [term]: [formatting/usage rule]

## Channel Adaptation
| Channel | Formality | Humor | CTA Style |
|---------|-----------|-------|-----------|
| Website | [adjust] | [adjust] | [style] |
| Email | [adjust] | [adjust] | [style] |
| Social | [adjust] | [adjust] | [style] |
| Support | [adjust] | [adjust] | [style] |
| Error messages | [adjust] | [adjust] | [style] |

## Sample Rewrites
### Homepage Hero
- **Before:** "[off-brand]" / **After:** "[on-brand]"
### Welcome Email
- **Before:** "[off-brand]" / **After:** "[on-brand]"
### Error Message
- **Before:** "[off-brand]" / **After:** "[on-brand]"
### Social Post
- **Before:** "[off-brand]" / **After:** "[on-brand]"

## Voice Checklist
- Does this sound like us?
- Could a competitor have written this?
- Would I say this to a customer?
- Does this pass the "but not" test?

Reference

For detailed frameworks on tone dimensions and archetype theory: see references/tone-dimensions.md


Related Skills

  • copywriting — use after this guide is set to write on-brand copy
  • french-copywriting — French-language copy, informed by this voice guide
  • social-copywriting — LinkedIn and X posts, adapted from this voice guide

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.