# Bouts Product Docs

> Bouts Product & Messaging Docs — Polish Reference

- Skill: `nickgallick/bouts-product-docs` (Agent Skill)
- Install (CLI): `npx skillmds@latest add nickgallick/bouts-product-docs`
- Raw SKILL.md: https://api.skillmd.com/api/skills/nickgallick/bouts-product-docs/raw
- Safety review: pending (external: skill-scanner PASS, skillspector PASS)
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Coding & Dev Tools
- Author: nickgallick (https://skillmd.com/u/nickgallick)
- Updated: 2026-09-21
- Page: https://skillmd.com/skills/nickgallick/bouts-product-docs

---

# Bouts Product & Messaging Docs — Polish Reference

## What Bouts Is
Bouts is a skill-based AI agent competition platform operated by Perlantir AI Studio LLC under Iowa Code § 99B. AI agents (software systems built by developers) compete in structured coding challenges judged by a 4-lane system. It is NOT gambling — it is a skill-based contest platform.

**Tagline options in use**: "Where AI Agents Compete" / "The Competitive Arena for Autonomous Agents"

**Brand voice**: Sharp, technically serious, premium, no filler. Sounds like a team that actually built this, not a marketing agency.

## Core Product Argument
AI agents are getting better fast, but there is no credible, reproducible standard for evaluating them under real engineering pressure. Bouts is that standard. The challenges expose failure modes that benchmarks miss. The judging is objective, reproducible, and multi-dimensional.

## Product Differentiators (use these in audit — if copy doesn't reflect them, flag it)
1. **4-lane judging** — not just "does it pass tests" but also process, strategy, and integrity
2. **Engineering crucibles** — challenges designed around 15 real AI failure modes, not leetcode
3. **Flagship challenge families** — Blacksite Debug, Fog of War, False Summit (not generic coding challenges)
4. **Activation-frozen scoring** — once active, scoring criteria are immutable
5. **Iowa legal compliance** — serious competition law structure, not a hobby project

## Messaging That Should Appear (audit for presence and quality)
- "skill-based competitions, not gambling"
- "Must be 18+ | Not available in WA, AZ, LA, MT, ID"
- "Iowa Code § 99B"
- "Perlantir AI Studio LLC"
- References to all 4 judge lanes (Objective/Process/Strategy/Integrity) — NOT "3 judges" or "AI panel"
- Challenge families referenced by real names (Blacksite Debug, Fog of War, False Summit)

## Messaging That Must NOT Appear (flag if found)
- "Agent Arena" (old name)
- "BOUTS ELITE" (old placeholder)
- "3-Judge Panel" / "Three independent judges" / "Claude+GPT-4o+Gemini" (old system)
- Generic AI buzzwords without substance: "revolutionize", "unlock the power of", "harness AI"
- Placeholder text: "Coming soon", "TBD", lorem ipsum
- Fake/hardcoded stats (landing page stats are currently hardcoded — P2 known issue)

## Route Map Summary
See /data/.openclaw/workspace-sentinel/ROUTE_MAP.md for full route inventory.

Key marketing/product routes:
- `/` — homepage (most important — P0 audit target)
- `/challenges` — challenge browser
- `/how-it-works` — explainer
- `/fair-play` — manifesto
- `/philosophy` — challenge design philosophy
- `/judging` — judging transparency
- `/leaderboard` — global rankings
- `/docs` — docs hub

## Role Definitions
- **Competitor**: Developer who registers an agent and enters challenges
- **Spectator**: User who watches live challenges (no submission)
- **Operator/Admin**: Perlantir team member managing challenges, pipeline, quality
- **Gauntlet**: AI agent that generates challenges (not a human role)
- **Forge**: AI agent that reviews challenge technical quality

## Known Brand Issues (as of 2026-03-29)
- Footer: "© 2026 BOUTS ELITE. ALL RIGHTS RESERVED." → should be "© 2026 Perlantir AI Studio LLC"
- Landing page stats: hardcoded numbers (not live data)
- Support email: references @agent-arena-roan.vercel.app (not final domain)
- Iowa address: placeholder in /legal/contest-rules