奇创岛

From «Atomic Habits»

Build an AI Habit Coach Toolkit Based on the Four-Step Loop

You'll distill the four-step habit loop, the habit stacking formula, and the 'never miss twice' principle from Atomic Habits into a reusable set of AI prompts — so you can call on the book's frameworks instantly whenever you start a new habit, without flipping through pages or relying on willpower.

Final work

An 'AI Habit Coach Prompt Toolkit'

Estimated time

1–2 hr

Submitted

…

Your final work

Purpose:Package the core frameworks of Atomic Habits into AI prompts you can grab at any moment — so every time you start a habit, hit a wall, or need to restart after a break, you can summon the book's wisdom in under 30 seconds instead of re-reading or grinding through on willpower alone.

Parts:

  • A four-step loop analysis prompt (System Prompt + user input template)
  • A habit stacking setup assistant prompt
  • A failure-restart diagnosis prompt
  • A 30-day review and identity confirmation prompt
  • An environment redesign suggestion prompt
  • A toolkit usage guide (which prompt to use when + notes)

Use cases:

  • · Paste directly into ChatGPT, Claude, Kimi, or any chat interface for immediate use
  • · Save as a personal habit-coach workflow for repeated reuse
  • · Share with friends or teammates to help them launch their own habit systems

Pick a topic

Pick the topic closest to you, or write a custom one when you submit.

Personal Life

Learning / Growth

Work / Projects

Tools you'll use from the book

Four-Step Loop System Prompt

Packages the four analytical dimensions — cue, craving, response, reward — into an AI role definition so that every habit conversation follows the loop structure automatically.

How to use it here:

Use this as the core System Prompt for your toolkit: tell AI upfront that 'you are a habit coach grounded in the four-step loop from Atomic Habits,' so all subsequent conversations stay within that framework without you having to re-explain the theory each time.

Boundaries:

Don't let AI make decisions for you. The four-step loop is a diagnostic framework — the final habit design must come from your real-life context.

Four Laws Diagnostic Prompt

Turns the four laws — make it obvious, attractive, easy, and satisfying — into a structured diagnostic template that checks your habit design for weak spots one law at a time.

How to use it here:

Before launching a new habit or after repeated failures, fill in your current habit and let AI score it against each of the four laws with specific improvement suggestions — tackle the weakest law first.

Boundaries:

Diagnostic results are reference suggestions, not absolute verdicts. The same habit may score differently depending on each person's environment and lifestyle.

Failure-Restart Prompt

A dedicated prompt designed for the most common scenario — a broken habit streak — that helps you distinguish between a one-off slip and a systemic issue so you don't abandon everything after one miss.

How to use it here:

Paste it in whenever your habit breaks for more than 2 days. Input the context and your state of mind at the time, and AI will categorize it as a 'slip vs. system bug' and give you the smallest viable restart plan — the AI-powered version of the book's 'never miss twice' principle.

Boundaries:

Don't use this to justify infinite restarts. Every restart should come with an updated execution plan, not just emotional reassurance.

Identity Vote Tracker

Translates 'every action is a vote for your identity' into a recordable, countable prompt tool so you can see the numeric trail of your identity accumulation.

How to use it here:

Each weekend, input which target habits you completed that week, and AI outputs them in the format 'You cast N votes for the person who [identity statement]' — reinforcing self-perception at the identity layer and giving your persistence a concrete achievement anchor.

Boundaries:

Identity votes are a motivation mechanism, not a performance metric. Don't get discouraged by a low count — focus on the trend, not the absolute number.

Habit Stacking Setup Assistant

Based on the 'After X I will Y' formula, helps you find the best insertion points from your existing daily anchors and automatically generates multiple stacking options to choose from.

How to use it here:

Input the 3–5 things you already do every day, and AI will list all usable anchors in chronological order, then recommend 2–3 optimal stacking positions for your new habit with reasoning — copy the output directly as your implementation intention.

Boundaries:

Stacking plans need real-world validation. AI-recommended anchors may not match your actual daily rhythm — test a stacking option for one week before committing to it.

Work rules

Your work MUST include

  • Include at least 3 complete, ready-to-use prompts (each with a System Prompt or full user template)
  • Label every prompt with 'when to use / what to input / expected output'
  • At least 1 prompt must directly address a real habit you are currently working on
  • Include a toolkit usage guide explaining the recommended order or context for each prompt
  • Reflect at least 2 core concepts from Atomic Habits (four-step loop / four laws / habit stacking / identity change, etc.)

Your work CANNOT just be

  • Don't just copy quotes or theory summaries from the book
  • Don't submit prompts you have never actually tested — you must note test results or revision records
  • Don't let AI fabricate your habit-execution experiences for you
  • Don't submit a single generic 'habit assistant' prompt with no scenario-specific tools
  • Don't let the toolkit stay on paper — at least one prompt must have been used in practice before submission

AI can help you here

Round 1: Help me identify the core scenarios my toolkit should address

When to use: You're just getting started and aren't sure which prompts to design, or you don't know which habit problem to tackle first.

I'm working through the '{{route name}}' project using *{{book title}}*, and my goal is to build a reusable AI habit coach toolkit.

Based on my habit situation below, help me identify 3–5 usage scenarios worth designing into prompts.

My habit situation:
[Describe the habits you've been trying or repeatedly failing at, and the moments when things most often go wrong]

Requirements:
1. For each scenario, explain 'at which moment AI intervention is needed'
2. Note which core concept from Atomic Habits each scenario maps to
3. Identify the highest-priority scenario (most frequent, most painful)
4. Suggest which prompt I should build first

Output:
- A list of 3–5 scenarios suited to my situation
- The 'trigger moment + corresponding book concept' for each
- A priority-order recommendation

Yellow placeholders need you to fill in before using the AI.

AI can help you organize ideas, but cannot make final judgments for you. Don't let AI fabricate experiences, cases, or misleading content.

Round 2: Help me turn the book's frameworks into usable prompt templates

When to use: You've decided which scenarios to build prompts for, but you're not sure how to translate the specific concepts from Atomic Habits into a well-structured prompt.

I'm working through the '{{route name}}' project using *{{book title}}*.

My chosen topic is:
{{chosen topic}}

The prompt scenario I want to design:
[Describe the 1–2 prompt names you have in mind, what role you want AI to play, and what you want it to do]

Please help me translate the following Atomic Habits frameworks into prompt templates:
1. Four-step loop (cue → craving → response → reward) for diagnosing habit problems
2. Four laws (make it obvious / attractive / easy / satisfying) for identifying design flaws
3. Habit stacking formula 'After X I will Y' for designing anchor points
(Pick the frameworks that fit your scenario — you don't need to use all three)

Requirements:
1. Each prompt must include a System Prompt section (telling AI its role and the framework)
2. Include a user input template (what the user fills in each time they use it)
3. Include an expected output format
4. Use real concepts from the book — don't invent new frameworks

Output:
- A complete, copy-ready prompt template
- Usage notes (when to use / what to input / things to watch out for)
- Potential pitfalls or edge cases (situations where this prompt may not apply)

Yellow placeholders need you to fill in before using the AI.

AI can help you organize ideas, but cannot make final judgments for you. Don't let AI fabricate experiences, cases, or misleading content.

Round 3: Help me check my toolkit for completeness and usability

When to use: You've written a first draft of your toolkit and want to confirm it's genuinely useful and nothing is missing before you submit.

I'm submitting a project work for the Shufang Island book project.

Book: *{{book title}}*
Project route: {{route name}}
My chosen topic: {{chosen topic}}

My work draft (AI Habit Coach Toolkit):
{{work draft}}

Please review this toolkit from the following angles:

1. Completeness: Does it cover the key stages of the habit lifecycle (launch / daily practice / restart after a break / periodic review)?
2. Usability: Does every prompt have a clear trigger moment and input requirement? Can a user understand 'when to use this' within 30 seconds?
3. Connection to the book: Does it reflect at least 2 core concepts from Atomic Habits? Is anything in the toolkit completely unrelated to the book?
4. Real-world testing: Is there any record of actual use, or does everything remain at the design stage?
5. Risk check: Are there any prompts that might lead AI to fabricate experiences for the user or give advice disconnected from their real context?

Output:
- Overall verdict (ready to submit / needs adjustment / needs a redo)
- What's working well
- Must-fix issues (with specific suggestions)
- Optional enhancements
- Suggested revised toolkit structure

Yellow placeholders need you to fill in before using the AI.

AI can help you organize ideas, but cannot make final judgments for you. Don't let AI fabricate experiences, cases, or misleading content.

Submit my work