✦ You asked for this
Free pack · $0A skill for Claude Code and Codex. Type /dalio and a decision. Five scouts argue it, and you get PROCEED, HOLD, or INVESTIGATE with a confidence score.
Take me to the installfree · one command✦ Get it
One line in your terminal downloads one file into your Claude Code and Codex skills folders. No git needed, just curl.
Your skills live somewhere else? The README on GitHub has the manual install path.
mkdir -p ~/.claude/skills/dalio ~/.codex/skills/dalio && curl -fsSL https://raw.githubusercontent.com/josephtandle/ray-dalio-council/main/dalio/SKILL.md -o ~/.claude/skills/dalio/SKILL.md && cp ~/.claude/skills/dalio/SKILL.md ~/.codex/skills/dalio/SKILL.md
✦ How it works
Five scouts with fixed lenses, a believability weighting system, and a synthesizer that produces one clean verdict. Modeled on how Ray Dalio ran decisions at Bridgewater.
Ray Dalio founded Bridgewater Associates in 1975. His book Principles lays out the idea behind this skill: believability weighting. Not all opinions deserve equal weight. The people with the strongest track record in a given domain should carry more influence over a decision than those without one. He also believed you should never trust your own thinking unchallenged. The best decisions come from stress-testing your position against people who disagree, each arguing their strongest case, before you commit. That is what this skill runs: five specialists, fixed lenses, weighted synthesis, one verdict.
| Scout | Weight | What it does |
|---|---|---|
| Optimist | w1.0 | Makes the strongest case FOR proceeding. Upside, momentum, and the cost of inaction. It never raises risks. That is not its job. |
| Skeptic | w1.0 | Makes the strongest case AGAINST. What are you missing? Which assumption is wrong? What would cause this to fail in 90 days? |
| Strategist | w1.2 | Evaluates long-term alignment. Is this the right move at the right time? Does it serve your core goals 6 to 12 months out? |
| Risk Analyst | w1.1 | Identifies specific failure modes and worst-case scenarios. For each risk: how likely is it, and what is the blast radius if it hits? |
| Domain Expert | w1.3 | Uses any context you provide to evaluate the decision from the ground level. If no context is given, it applies general business principles. |
✦ Use it now
In Claude Code, type /dalio and your question. In Codex, type $dalio instead. Start with the plain question. Add context when you want the Domain Expert to use it.
Five scouts argue it with no context needed. Works for any decision.
/dalio Should we launch the referral program this week?
The Domain Expert uses your context. The other four still argue their fixed lens.
/dalio context: "bootstrapped SaaS, 150 users, $5k MRR" -- Should we hire a salesperson?
✦ Get unstuck
Check that the file is at ~/.claude/skills/dalio/SKILL.md for Claude Code, or ~/.codex/skills/dalio/SKILL.md for Codex. If your skills live somewhere else, the README on GitHub has the manual install path.
The same command installs both. In Codex, type $dalio instead of /dalio.
You do not need it. The command uses curl to download one file.
Ask the bare question. If no context is given, the Domain Expert applies general business principles.
SPLIT means a close vote or low confidence. The synthesizer names the dissent so you can see who disagreed and why.
HOLD often comes with a condition, like "Proceed when X is resolved." INVESTIGATE lists exactly what you need to find out before you decide.
Email joe@mastermindshq.business. Say which step you were on and paste what you saw.
✦ What next · paid · optional

You came for one piece of my Claude Code setup. If you have your own website, the next useful job for that setup is getting the site found.
Get your website listed on Google and Bing so people (and ChatGPT) can find you by your name and your service. ChatGPT finds websites mostly through Bing, so we tell Bing about you too. Your own Claude Code does the work on your computer, one prompt at a time.
Not for you? Close this page. Everything above stays free.
Created by Joe Che.
ChatGPT is a trademark of OpenAI. Claude is a trademark of Anthropic. This page is not affiliated with either.