# Brand Voice For Bouts

> The specific voice, tone, writing rules, approved vocabulary, and forbidden phrases for all Bouts content. Use when writing or reviewing any Bouts copy — landing pages, social posts, emails, reports, DMs, press pitches — to ensure the brand sounds authoritative, data-driven, and impossible to confuse with generic AI marketing.

- Skill: `nickgallick/brand-voice-for-bouts` (Agent Skill)
- Install (CLI): `npx skillmds add nickgallick/brand-voice-for-bouts`
- Raw SKILL.md: https://api.skillmd.com/api/skills/nickgallick/brand-voice-for-bouts/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Marketing & Growth
- Author: nickgallick (https://skillmd.com/u/nickgallick)
- Updated: 2026-09-08
- Page: https://skillmd.com/skills/nickgallick/brand-voice-for-bouts

---


# Brand Voice for Bouts

## Voice attributes
- **Authoritative but not arrogant:** Show the data. Let it speak. Do not claim to be the only benchmark that matters.
- **Technical but accessible:** A developer understands every blog post. A non-developer understands every tweet. Same data, different depth.
- **Competitive but fair:** Celebrate great agent performance. Never mock poor performance. Analyze, do not ridicule.
- **Data-driven, never hype:** Every claim has a number. "Agents improved" → "Agents improved 18% on Recovery scores this month."
- **Insider language:** Use platform terms naturally — Blacksite Debug, CDI, Recovery Judge, Versus, Abyss, Integrity Judge. This builds brand vocabulary.

## What we NEVER say
- "The best AI benchmark"
- "Unbeatable" / "revolutionary" / "game-changing"
- "Our AI is smarter than X"
- "Guaranteed results"
- Any vague superlative without a data point

## What we ALWAYS include
- Specific numbers: "23% solve rate," "379-point ELO spread," "15 failure archetypes"
- Attribution: "According to this week's challenge data…" / "The Bouts AI Agent Index shows…"
- Methodology references: "Scored across 5 independent judge lanes…" / "Using contamination-resistant challenge generation…"

## Writing rules
- Lead with the insight, not the setup
- One idea per paragraph, one point per tweet
- Use Bouts terminology naturally
- Include a data point in every post
- End with a question or call-to-action when appropriate
- Short sentences. Active voice.

## Sample rewrites

**Wrong:**
"Our amazing AI platform helps developers understand their agent performance in a revolutionary new way."

**Right:**
"Bouts scores AI agents across 5 independent judge lanes — not just did it work, but how it worked and whether you'd trust it in production."

---

**Wrong:**
"We're the best AI benchmark available today."

**Right:**
"Static benchmarks cluster top models within 5-8%. Bouts produces 379-point ELO spreads. The difference is what you measure."

---

**Wrong:**
"Check out our new challenge!"

**Right:**
"This week's Blacksite Debug has a 12% solve rate so far. Here's what the 88% are getting wrong."