Skip to content

BrianRWagner/ai-marketing-claude-code-skills / brand-voice-extractor

brand-voice-extractor

Verified

Extract or build a distinct brand voice profile that AI agents can use to produce on-brand content every time. Two modes: Extract (analyze content you're proud of) or Build (construct a voice from scratch). Outputs a complete voice profile with personality traits, tone spectrum, vocabulary guide, rhythm patterns, and example phrases. Use this before any content creation skill for consistent, human-sounding output.

FileKindSize
README.mdmarkdown2.8 KB
SKILL.mdskill md11.7 KB
references/examples.mdmarkdown7.6 KB
sample-output.mdmarkdown5.4 KB
scripts/analyze-voice.mjsscript6.6 KB

sample-output.md

View on GitHub
# Sample Output: Brand Voice Extractor

*This is a real example of the kind of voice profile the skill produces. Extract mode, sourced from LinkedIn posts, emails, and website copy.*

---

## Inputs Analyzed

5 pieces of content:
- Homepage copy + About page
- 3 top-performing LinkedIn posts (selected by engagement)
- Welcome email to new subscribers

---

## Output: Voice Profile

---

# Alex Chen — Voice Profile

## Voice Summary
Sounds like the engineer who learned to explain things in English — sharp, specific, and allergic to vague claims. Confident enough to say what others won't, grounded enough to show the math. The tone communicates: *I've seen your problem before, here's exactly what it is, and here's what you actually do about it.* No corporate wrapper. No hand-holding. Genuine expertise, human delivery.

## Core Personality Traits

- **Evidence-first:** Never makes a claim without a number or example attached. "Most companies have this problem" becomes "73% of B2B companies we've audited have this problem." Credibility is built on specifics, not assertions.
- **Impatient with vague:** Allergic to empty language. "Leverage synergies" = instant credibility drain. Explains complex things simply because it's harder, not easier, to do that well.
- **Peer-level honesty:** Doesn't talk down or up. Talks across. The stance is: I'm a bit further down the same path you're on, here's what I've learned.
- **Systems thinker:** Sees problems as broken systems, not broken people. Solutions are always structural, not motivational.

## Tone Spectrum

| Dimension | Position | Notes |
|-----------|----------|-------|
| Formal ↔ Casual | Casual-professional | Conversational rhythm, no slang, occasional contractions |
| Serious ↔ Playful | Serious with occasional wit | Humor earns its place — never decoration |
| Reserved ↔ Bold | Bold | Strong claims delivered calmly, not aggressively |
| Simple ↔ Sophisticated | Simple words, complex ideas | Short words, dense thinking |
| Warm ↔ Direct | Direct with warmth | Directness is respect — doesn't waste the reader's time |

## Vocabulary

**Words/phrases to USE:**
- "Here's the thing..." — setup for real talk
- "What that actually means..." — lands the insight
- "We've seen this pattern..." — authority through exposure
- Specific numbers always: "83% of accounts," "6 weeks," "$40K deal"
- "The fix is structural, not motivational" — systems framing
- "Here's what you actually do..." — permission to act

**Words/phrases to AVOID:**
- "Leverage" (as noun), "synergy," "robust," "comprehensive"
- "I think" or "maybe" in authority zones — kills credibility
- "Passionate about," "excited to share" — performative
- "In today's fast-paced world" — instant trust drain
- "Just" as minimizer ("just do X") — dismisses the difficulty
- "Game-changing," "revolutionary," "cutting-edge"

**Jargon level:** Light. Uses technical terms (CAC, LTV, ARR) but translates them on first use.

**Profanity:** Rare. Only when it adds genuine emphasis, never as a habit.

## Rhythm & Structure

**Sentences:** Mix of 3-5 word punches and 10-15 word explanations. Rarely longer than 20 words.

**Paragraphs:** 1-2 sentences. White space is structural, not decorative.

**Openings (signature moves):**
- Contrarian claim: "Most sales coaching focuses on the wrong variable."
- End result first: "I audited 12 outbound sequences last quarter. 10 had the same fatal flaw."
- Problem reframe: "You don't have a lead quality problem. You have a qualification problem."
- Pattern reveal: "Every time I see a pipeline stall like this, it's the same root cause."

**Formatting:** Clean. Headers for long-form. Bold for emphasis, not decoration. Lists when genuinely listing things, not to break up paragraphs. Minimal emoji.

## POV & Address

**First person:** I (never "we" unless referring to a team)
**Reader address:** You, direct
**Relationship stance:** Peer who's slightly ahead. "I was where you are six months ago."

## Example Phrases

**On-brand:**
- "Your pipeline isn't stalling because of follow-up cadence. It's stalling because buyers can't explain your value to their boss."
- "I ran the same diagnostic on three clients in Q4. Different industries, same broken step."
- "The math on this is simple. Most teams just don't do it."

**Off-brand:**
- "We're passionate about helping companies achieve their sales goals through innovative, cutting-edge methodologies." — Vague, corporate, zero specifics
- "I'm so excited to share this game-changing framework with you all!" — Performative, no substance
- "There are many factors to consider when evaluating your pipeline health, and it's important to think about..." — Hedging, no point

## Do's and Don'ts

**DO:**
- Lead with a specific observation or result — never a generic intro
- Attach a real number to every claim that can carry one
- Reframe the problem before offering the solution
- Keep paragraphs to 1-2 sentences — always
- Write the last line like it's the only line someone will read

**DON'T:**
- Start with "I" as the first word of a post
- Use "passionate about" or "excited to share"
- Pad sentences with qualifiers when you're sure of something
- Write paragraphs longer than 3 sentences
- Close without a clear landing (point, action, or implication)

---

*Generated by Brand Voice Extractor | Free in the [AI Marketing Skills](https://github.com/BrianRWagner/ai-marketing-claude-code-skills) library*
reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis.Stars: 472kUpdated Claude CodeCodexVerifiedInstall
diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness.Stars: 323kUpdated Claude CodeCodexVerifiedInstall
dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependenciesStars: 323kUpdated Claude CodeCodexVerifiedInstall
executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is availableStars: 323kUpdated Claude CodeCodexVerifiedInstall