Skip to content
Productivity
Command

/reality-test

Counter confirmation bias by actively seeking evidence that could prove you wrong. Test assumptions before committing.

BOOST
From plugin
neurofoo-agent-skills
12228 skills28 commands
Install
> /plugin marketplace add neurofoo/agent-skills

How it fires

How this command gets triggered: by you, by Claude, or both.

  • Fires itselfClaude auto-loads it when your prompt matches the work.
  • You can call itInvoke it directly when you want it.
  • Slash command/reality-test

Context preview

What this command does when you run it.

Counter confirmation bias by actively seeking evidence that could prove you wrong. Test assumptions before committing.

Command definition

reality-test.md

Reality-Test Assumptions

Counter confirmation bias by actively seeking evidence that could prove you wrong. Test assumptions before committing.

Instructions

Identify key assumptions underlying your decision and design ways to test them with real data, not just analysis.

Output Format

**Decision/Belief Being Tested**: [What we're evaluating]

---

Identify Assumptions

**What are you assuming to be true?**

| Assumption | Confidence | Importance | |------------|------------|------------| | [assumption 1] | High/Med/Low | High/Med/Low | | [assumption 2] | High/Med/Low | High/Med/Low | | [assumption 3] | High/Med/Low | High/Med/Low |

---

Consider the Opposite

For each key assumption, what would convince you it's wrong?

| Assumption | What Would Change My Mind | |------------|--------------------------| | [assumption] | [disconfirming evidence] | | [assumption] | [disconfirming evidence] |

**Deliberately seek disconfirming evidence:** Where can we look for evidence that we're wrong?

  • [Source 1]
  • [Source 2]

---

Base Rates

**What's the general success rate in situations like this?**

| Situation | Base Rate | Source | |-----------|-----------|--------| | [similar situation] | [%] | [where we know this from] |

**Are we assuming we're different from the average?** [Yes/No] If yes, what specifically makes us different?

---

Ooch: Small Bets

**Can we test this with a small experiment before fully committing?**

| What to Test | The "Ooch" | What We'd Learn | Cost/Time | |--------------|-----------|-----------------|-----------| | [assumption] | [small test] | [insight] | [cost] | | [assumption] | [small test] | [insight] | [cost] |

**Pilot possibility**: [How to test at small scale first]

---

Ask the Experts

**Who has direct experience with this?**

| Expert/Source | What They'd Know | How to Reach Them | |---------------|------------------|-------------------| | [who] | [their expertise] | [how to contact] |

**What would a skeptic say?** > [The best argument against our current view]

---

Evidence Summary

| Assumption | Evidence For | Evidence Against | Verdict | |------------|--------------|------------------|---------| | [assumption] | [supporting] | [contradicting] | [likely true/uncertain/likely false] |

---

Revised Confidence

**Before reality-testing**: [Initial confidence %] **After reality-testing**: [Revised confidence %] **Key change**: [What we learned]

**Next step**: [Action to take based on findings]

Guidelines

  • Seek out disagreement, not confirmation
  • "What would have to be true?" tests logic
  • "What usually happens?" tests base rates
  • "Can we try a small version?" tests reality
  • Talk to experts who've done it, not just thought about it

$ARGUMENTS

Read more
Ships withneurofoo-agent-skills

A collection of thinking frameworks and decision-making tools as skills for AI coding assistants.

Get the whole plugin
Stats
122
Stars
19
Forks
Quiet
Maintenance
MIT
License
9mo ago
Last commit
9mo ago
Created
23h ago
Added

Repo: neurofoo/agent-skills

Other commands on neurofoo-agent-skills.