scout-mindset-bias-check — Claude Code skill: what it does
scout-mindset-bias-check is a Claude Code skill that checks forecasts with Scout Mindset procedures, including when an estimate feels emotional or stuck at 50/50.
no license 119 installs ★ 203 updated 2026-09-01
npx skills add lyndonkl/claude --skill scout-mindset-bias-check -a claude-codeWhat scout-mindset-bias-check does
The skill runs a menu of procedures over a prediction you already have. There is a reversal test, a scope-sensitivity check, a status-quo-bias test, a confidence-interval audit, and a full bias audit that works through a checklist of distortions and ranks the fixes. One menu item points at three reference files in resources/: scout-vs-soldier.md, cognitive-bias-catalog.md, debiasing-techniques.md.
The reversal test takes five steps. State the call and the probability, list what supports it, turn each piece around, then ask whether the flipped version would still move you. Answer no, or hesitate, and the skill flags special pleading and advises nudging the number 10-15% toward 50%. Treat that as a crude correction. It does not tell you whether the forecast itself is any good.
Start with a number on the table. A vague hunch leaves the steps with nothing to chew on. After that it is plain back-and-forth in chat, and the skill wants your answers rather than a dataset.
What it helps with
- catch motivated reasoning inside your own estimate
- check whether a forecast leans toward the current state of affairs
- widen or narrow a confidence interval that was never tested
- unstick a prediction sitting at 50/50
- run a cognitive-bias checklist before handing over a forecast
Example prompts for the agent
- Run the reversal test on this: I think candidate A wins at 75%, based on polls, campaign funding and debate ratings.
- Check scope sensitivity for the project timeline at one month, ten months and a hundred months.
- Give me a full bias audit of my product launch forecast.
How to install scout-mindset-bias-check
1Skills CLI (Vercel)
npx skills add lyndonkl/claude --skill scout-mindset-bias-check -a claude-code2Manually: put the files in place
~/.claude/skills/scout-mindset-bias-check/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/scout-mindset-bias-check/
1Skills CLI (Vercel)
npx skills add lyndonkl/claude --skill scout-mindset-bias-check -a codex2Manually: put the files in place
.agents/skills/scout-mindset-bias-check/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add lyndonkl/claude --skill scout-mindset-bias-check -a cursor2Manually: put the files in place
.cursor/skills/scout-mindset-bias-check/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install lyndonkl/claude skills/scout-mindset-bias-check2Manually: put the files in place
.github/skills/scout-mindset-bias-check/SKILL.md1Skills CLI (Vercel)
npx skills add lyndonkl/claude --skill scout-mindset-bias-check -a gemini-cli2Platform CLI
gemini skills install https://github.com/lyndonkl/claude --path skills/scout-mindset-bias-checkManually: put the files in place
.devin/skills/scout-mindset-bias-check/SKILL.mdThe legacy .windsurf/skills/ path still works
Skills CLI (Vercel)
npx skills add lyndonkl/claude --skill scout-mindset-bias-check -a opencodeClaude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about scout-mindset-bias-check
How to install scout-mindset-bias-check in Claude Code?
Run npx skills add lyndonkl/claude --skill scout-mindset-bias-check -a claude-code. The skill file then sits at ~/.claude/skills/scout-mindset-bias-check/SKILL.md.
What does scout-mindset-bias-check do?
It checks a forecast for cognitive bias: reverses the supporting evidence, tests sensitivity to magnitude, looks for a pull toward the status quo, and inspects interval width. You get a ranked list of corrections at the end.
How is the full bias audit different from the reversal test?
The reversal test handles one conclusion and the evidence behind it. The full audit walks a checklist of the major distortions and ranks what to fix first, so it takes longer and covers more ground.