Brand Voice & Tone

Apply configurable brand voice to any written content. Reads brand identity from a config file and adapts voice pillars, vocabulary, tone calibration, and writing patterns accordingly. [EXPLICIT]

Brand Configuration

Search for brand voice config in this order:

1. Path passed as argument
2. ./brand-voice.json in working directory
3. ./brand-config.json (voice section)
4. ~/.claude/brand-voice.json

Config Schema

{
  "brand": {
    "name": "YourBrand",
    "positioning": "Your positioning statement",
    "claim": "Your main tagline or claim",
    "claimVariants": [
      "Variant 1. Tagline.",
      "Variant 2. Tagline."
    ],
    "philosophy": "Core philosophy or methodology"
  },
  "voice": {
    "pillars": [
      {
        "name": "Pillar Name",
        "description": "What this pillar means",
        "guidelines": ["Guideline 1", "Guideline 2"]
      }
    ]
  },
  "vocabulary": {
    "preferred": {
      "generic term": "brand term",
      "deliverables": "results"
    },
    "avoid": ["synergy", "leverage", "seamless"]
  },
  "toneCalibration": {
    "proposals": "confident + structured",
    "marketing": "warm + energetic",
    "technical": "clear + precise",
    "internal": "friendly + direct",
    "crisis": "calm + accountable"
  },
  "closingLines": {
    "external": ["Closing line 1", "Closing line 2"],
    "proposals": ["Proposal closing 1"],
    "internal": ["Internal closing 1"]
  }
}

Default (No Config)

If no config found, use professional neutral voice: clear, direct, active voice, no jargon, human-centered. Prompt user to provide brand config for branded output. [EXPLICIT]

Voice Pillars

Read from voice.pillars in config. Default 4-pillar framework:

1. Close & Human

Companion, not vendor. Warm, direct, personal. Talk TO people, not AT them. Use "we" and "you" naturally. [EXPLICIT]

2. Confident & Grounded

Lead with facts, not adjectives. Active voice. Confidence without arrogance: "We deliver" not "We are proud to deliver."

3. Forward-Looking & Optimistic

Focus on progress, not perfection. Frame challenges as opportunities. End on a forward note. [EXPLICIT]

4. Simple & Direct

Short sentences. One idea per sentence. Avoid jargon when plain language works. Active verbs: build, deliver, transform, improve. [EXPLICIT]

Writing by Content Type

Headlines & Titles

Short, punchy, sentence case. Lead with action or aspiration. Optionally highlight one key word in brand primary color. [EXPLICIT]

Body Copy

2-4 sentences per block. Active voice. End with forward-looking statement or CTA. Subtly reference brand philosophy. [EXPLICIT]

Proposals & Reports

Open with client's challenge, not credentials. Data-anchor claims. Structure: Context > Approach > Expected Impact. Close with warm, direct invitation. [EXPLICIT]

Social / LinkedIn

Start with human hook. Conversational, no buzzwords. Concrete insight or takeaway. End with tagline when appropriate. [EXPLICIT]

Emails (External)

Clear subject (benefit or question). Reference specific context. One key message. Single CTA. Warm professional sign-off. [EXPLICIT]

Slide Text

6-10 words max per headline. No full paragraphs. "Eyebrow + headline" structure. Tagline on closing slides. [EXPLICIT]

Internal Communications

Same warmth, slightly more informal. Reference team identity. Celebrate small wins, frame feedback as improvement. [EXPLICIT]

Language Rules

Do use: We, our, you, your. Build, deliver, transform, improve, create, evolve. Sentence case for headlines.

Avoid: Synergy, leverage, seamless, cutting-edge, world-class, innovative (unless earned). Passive voice. Exclamation marks in professional contexts. ALL CAPS. Vague superlatives without proof.

Read additional rules from vocabulary.preferred and vocabulary.avoid in config. [EXPLICIT]

Tone Calibration

Read from toneCalibration in config. Default matrix:

Context	Tone	Notes
Corporate proposals	Confident + structured	Facts-forward, evidence-based
Marketing / social	Warm + energetic	Human stories, optimism
Recruiting / talent	Inspiring + honest	Growth mindset
Technical docs	Clear + precise	Plain language
Events / launches	Bold + celebratory	Claims front and center
Internal	Friendly + direct	Informal, motivating
Crisis / issues	Calm + accountable	Transparent, solution-focused

Claim & Tagline System

Read from brand.claim and brand.claimVariants. [EXPLICIT]

Rules:

• Claims work as standalone statements
• Extended variants follow the pattern: "[Adjective] [noun]. [Separator]."
• Positioning tag works as sign-off or label

Manifesto Tone Pattern

For high-impact copy, use anaphoric structure from brand philosophy:

Every [moment/person/project/step] [is/has/connects]. [EXPLICIT]
That's why we [action]. [EXPLICIT]

Assumptions & Limits

• Voice config is optional; skill works with defaults but output is generic
• Cannot invent brand identity — needs config or user-provided brand context
• Tone calibration is a dial, not a switch — content types blend tones
• Vocabulary mapping covers common substitutions; domain-specific terms need explicit config
• Not a copywriting skill — provides voice guidelines; actual copy requires creative judgment

Edge Cases

• No brand config: Use neutral professional defaults; note output is unbranded
• Conflicting tone context (crisis + marketing): Crisis tone always wins in blended contexts
• Non-English content: Apply same voice pillars; adapt vocabulary to target language idioms
• Multiple brands: Accept brand config path as argument; switch per invocation

Validation Gate

• Brand config loaded (or defaults applied with note)
• Voice pillars applied consistently throughout content
• Vocabulary uses preferred terms, avoids banned terms
• Tone matches content type calibration
• Active voice dominant (< 15% passive)
• Headlines in sentence case
• Closing line appropriate to context
• No jargon without explanation

---

Author: Javier Montano | Last updated: March 18, 2026

Usage

Example invocations:

• "/brand-voice" — Run the full brand voice workflow
• "brand voice on this project" — Apply to current context