Free Claude Skill · v1.0

Pressure-test any strategy through the Playing to Win cascade.

A Claude Code skill that applies Roger Martin and A.G. Lafley's choice cascade to your idea, offer, or product. It scans your codebase, runs market research, and produces a sharpened strategy with a clear verdict: Proceed, Narrow, Test manually, Park, or Kill.

What it does

You ask Claude to "build me a Playing to Win framework around this project" or "pressure-test this idea." The skill takes over.

If you invoke it inside a codebase, it reads your README, package files, route tree, marketing copy, pricing config, and analytics modules to extract your current customer, positioning, offer, and capabilities. It runs market research through whatever web tools are available, labels facts vs inferences vs assumptions, then produces a 12-section Standard Output:

  • Snapshot, Verdict, and Confidence
  • Winning Aspiration and Where to Play
  • How to Win, Required Capabilities, Management Systems
  • What Must Be True and the Red-Team Pre-Mortem
  • Sharpened Version, 30-Day Validation Plan, Presentation-Ready Summary

It ends with a single recommendation: Proceed, Narrow, Test manually, Park, or Kill. No fluff. No fake encouragement. The voice is direct, skeptical, and practical.

Who it's for

  • Founders evaluating whether to ship, narrow, or kill an idea
  • Operators inheriting a product or business unit and asking where to play
  • Leadership teams that need a real choice cascade, not another deck
  • Solo builders who want a tool that argues back instead of agreeing

It assumes you already know the framework. It applies it. It does not teach it.

Install

Download the zip, unpack it, and drop the playing-to-win/ folder into your skills directory.

# Claude Code, user-level (available across all projects) unzip playing-to-win.zip -d ~/.claude/skills/ # Claude Code, project-level (scoped to this repo only) unzip playing-to-win.zip -d ./.claude/skills/ # Codex unzip playing-to-win.zip -d ~/.agents/skills/

Restart your Claude Code session if it was already running. Then invoke it inside any project or conversation:

  • "Build me a Playing to Win framework around this project"
  • "Pressure-test this idea: [your idea]"
  • "Red-team this strategy"
  • "Where should we play with [offer]?"
  • "P2W cascade for [X]"

What's inside

playing-to-win/ ├── SKILL.md # Operator manual + 12-section output contract ├── README.md # Install and usage ├── TEST_PLAN.md # Trigger tests and structural checks └── references/ ├── 01_strategy_quality_rubric.md # Verdict scale, quality gates, red flags ├── 02_strategy_example_bank.md # Weak / strong / sharpened calibration ├── 03_assumption_test_bank.md # Practical tests by assumption type ├── 04_positioning_patterns.md # Strong / weak positioning + wedges ├── 05_strategy_preferences.md # Output style and tone calibration └── 06_execution_patterns.md # Rollout and adoption patterns

Why I built this

I rolled out Playing to Win across twelve teams at Verizon. All twelve hit their goals. It is the only strategy framework I have ever seen that survives contact with a P&L, a quota, and a skeptical engineer all at the same time.

The problem with most strategy work is that it lives in slides. You write a deck, present it, and three months later nobody can remember the choices it asked you to make. This skill turns the cascade into a callable function. You can pressure-test an idea on a Tuesday morning, get a sharpened version by Tuesday afternoon, and know whether to proceed by Wednesday.

Operator-built. Free. Adapt and improve.

Attribution

Playing to Win is the strategy framework developed by Roger L. Martin and A.G. Lafley, documented in Playing to Win: How Strategy Really Works (Harvard Business Review Press, 2013). This skill applies their cascade. It does not replace the book. Buy the book.

If you find this useful, the Field Notes newsletter is where I write up the operator lessons behind tools like this one. Free, occasional, no pitches.