── ── Mental model
Checklist
A checklist is a short written list of must-not-skip steps for a complex, repeated operation. It externalizes working memory — which fails under load, ego, time pressure, and fatigue — so attention stays on the situation instead of step recall. From Boeing in 1935 to Pronovost's ICU lines to the WHO surgical checklist, the documented effect is large, cheap, and repeatable.
Run Checklist on a real problem
Bring something you're actually deciding — free, in the browser.
How it works
Step 1 — Identify the operation: operation name, frequency, stakes of errors, current failure modes, near-miss history.
Step 2 — List 15-25 candidate items: steps that have been skipped, steps that prevent major errors, coordination steps, bias-catching steps, reversibility checkpoints.
Step 3 — Cut to 6-10: keep only items that are critical, easily missed by skilled operators, and verifiable (yes/no signal). Longer lists get truncated in use.
When to use it
- user says 'let's make sure we don't skip anything', 'we need a pre-flight', 'build me a checklist for X', 'we keep missing steps in our deployment / close / launch', 'Gawande / Pronovost / checklist manifesto'
When not to use it
the task is simple and one-off with no recurrence; the user is asking for a to-do list or project plan (not a must-not-skip operational gate).
Worked example
The AI-Agent Pre-Flight Checklist (2023–2026)
The checklist was born in 1935 because the B-17 became "too much airplane for one man to fly." Between 2023 and 2026, software crossed the same threshold: autonomous AI agents that write code, send emails, move money, and talk to customers became too much system for one deploy to eyeball. The industry's response recapitulated aviation's — quietly, and in exactly the shape this skill teaches. The pre-deployment routine that emerged across the ecosystem (model system cards, eval suites, staged rollouts, guardrail reviews) is a checklist…
Install this skill (free, MIT)
npx skills add deciqAI/knowledge-skillsUseful? Star the repo — stars help other builders find it.
FAQ
Why do checklists work if experts already know the steps?
Because failure isn't ignorance — it's ineptitude under load: known steps skipped when tired, rushed, or confident. The checklist moves step-recall out of working memory so expertise can focus on judgment. That's why they help experienced operators most, exactly the people who resist them.
What's the difference between DO-CONFIRM and READ-DO checklists?
READ-DO means read each step, then perform it — right for unfamiliar or emergency procedures. DO-CONFIRM means work from memory, then pause at a checkpoint to verify nothing was skipped — right for fluent routines. Choosing the wrong mode makes the checklist feel like bureaucracy.
What makes a checklist good rather than bureaucratic?
Brevity and killer-item focus: 5–9 items, only the steps that are both skippable and catastrophic, with clear pause points. A checklist that tries to encode the whole procedure gets abandoned; one that catches only the fatal omissions gets used for decades.
Related mental models
Most learning models treat acquisition as the final destination.
Working memory holds ~4 chunks of novel information — a hard limit.
Eight interconnected domains constitute human cognition: (1) Perception & Attention, (2) Memory, (3) Language & Thought, (4) Decision Making & Judgment, (5) Metacognition, (6) Emotion &…
Small commitments today dramatically raise the probability of large commitments tomorrow — people have a strong drive to behave consistently with prior choices, especially ones that…