Free Kit • On-Screen + Printable

AI Foundations Free Kit

Use AI without getting burned. This is a practical toolkit: task classification, steering scripts, week-1 reps, and reusable templates. Read it on-screen, copy what you need, or print clean sections.

Prerequisite for AI Blueprint™: This kit is the entry layer. It establishes the minimum mental model required before AI outputs can be governed.

Jump to Section 1 Templates Pack

Time: 20–40 min Output: reusable prompts + workflow map Mode: Foundations (pre-governance)

Core rule: AI produces candidates. Humans provide standards.

Navigate: Start 0 1 2 3 4 5 6 7 Templates

Start Here — Progress Scoreboard

Check any three and the kit did its job.

I classified my tasks correctly

I can tell AI-friendly vs hybrid vs human-essential.

I ran the 3-pass steering loop once

Draft → Critique → Rewrite.

I completed at least one Week-1 win

A 10–15 min rep that produced something useful.

I added a human constraint to an AI output

One “human add” that keeps judgment alive.

I know my next capability level

Level 1, 2, or 3 — and a safe mission to try.

I saved at least one template I’ll reuse

This becomes muscle memory, not a PDF I forget.

Go to Section 0 →

Section 0 — Orientation: What AI actually is

AI is a language prediction system. It’s fast at drafting and organizing — and it can sound confident even when it’s wrong.

Pocket rule: AI produces candidates. Humans provide standards.

Do This Now (2 min)

Pick an AI role before you prompt

Drafter (rough drafts)
Cleaner (clarity + structure)
Summarizer (compress)
Brainstormer (options)
Critic (find flaws)

Why this matters

Most “AI slop” is just role confusion. If you don’t tell AI what job it’s doing, it guesses. And guessing produces generic output.

Template AI Role Picker

Act as my [ROLE].
My goal is [OUTCOME].
My constraints are [CONSTRAINTS].
Before you start, ask me 2 clarifying questions.

Tip: If AI doesn’t ask questions, your prompt is too vague or your constraints are missing.

Section 1 — Task Classification Cheat Sheet

Most AI frustration comes from delegating the wrong layer of work.

Test 1

Irreversibility

If this ships wrong, can you undo it easily?

Test 2

Reputation

Would it damage trust if it’s wrong or sloppy?

Test 3

Judgment

Does it require taste, ethics, or accountability?

Rule: If 2+ tests are “yes,” it’s Hybrid or Human-Essential. AI can assist, not own.

AI-Friendly

Summaries
Rewrites for clarity
Outlines & checklists
Routine emails
Extract action items
Format tables

Hybrid

Marketing copy
Proposals
Sales sequences
Competitive analysis
Content calendars
Support macros (brand tone)

Human-Essential

Pricing decisions
Negotiations
Hiring/firing
Crisis responses
Legal/medical/financial conclusions
Final decisions

Do This Now (5 min) Classify your top 10 weekly tasks

Here are my 10 recurring weekly tasks:
[paste list]

Classify each as AI-Friendly, Hybrid, or Human-Essential.
Explain why using: irreversibility, reputation, judgment.
Then tell me the top 3 tasks I should delegate to AI first.

Pocket rule: If AI keeps disappointing you, you’re delegating the wrong layer.

Section 2 — The First Draft Fallacy

AI isn’t a vending machine. It’s iterative. If you stop after draft #1, you stop mid-process.

Example (fast)

Three versions of the same ask

Vague: “Write a professional email about a delay.”
Outcome: “Reassure client, explain 1-week delay, preserve trust.”
Steering: “Make it calm/accountable, not defensive. Under 150 words.”

What changed?

Not “prompt magic.” You supplied: outcome, audience expectations, and constraints. That’s steering — and steering is the skill.

Pocket rule: The first draft is a receipt. The second draft is the meal.

Template 3-Pass Steering Loop

PASS 1 — DRAFT
Draft [ASSET] for [OUTCOME].
Audience: [WHO].
Tone: [TONE].
Length: [LENGTH].

PASS 2 — CRITIQUE
Critique your draft.
List the top 5 weaknesses: clarity, tone, missing info, risk, persuasion.

PASS 3 — IMPROVE
Rewrite fixing the top 3 weaknesses.
Keep structure but tighten language.
Give 3 alternative opening lines.

Do This Now (5 min): Pick one small asset (email, caption, summary) and run the 3-pass loop once.

Section 3 — Capability Ladder

You don’t need every feature. You need the next safe step.

Level 1

Basic Assistance

Q&A
Summaries
Rewrites
Formatting

Safe mission: Notes → outline → clean rewrite.

Level 2

Multimodal

PDFs
Screenshots
Images

Safe mission: Extract key facts → summarize for non-experts.

Level 3

Chained Tasks (no autonomy)

Multi-step workflows
Checklists
Reusable templates

Safe mission: Workflow + checklist + template. No final decisions.

Pocket rule: Don’t climb levels for ego. Climb for leverage.

Do This Now (3 min) Pick your next level

I am currently comfortable at Level [1/2/3].
Give me one low-risk “mission” I can run this week in 10–15 minutes.
It must produce a reusable output (checklist, template, or summary).
Ask me 1 clarifying question before you answer.

Section 4 — Week-1 Wins (Mini Program)

Do one rep per day. Stop early if you get value.

7-Day Plan

Day 1: Clean messy notes
Day 2: Draft a routine email
Day 3: Create a checklist
Day 4: Rewrite a document
Day 5: Extract action items
Day 6: Draft a simple SOP
Day 7: Combine into a template pack

What “good” looks like

10–15 minutes
One usable output
Saved as a template
Repeatable next time

Pocket rule: Confidence comes from reps, not reading.

Prompt Pack Week-1 Win (universal)

I’m doing a 10–15 minute “Week-1 Win.”
Input: [paste notes / email context / messy doc]
Goal: [what I want at the end]
Constraints: [tone/length/format]

1) Produce the output.
2) Then propose 2 ways to make it more reusable as a template.

Section 5 — Skill Anxiety (Quick Reframe)

AI doesn’t make you dumb. Misusing it does.

Two-Lane Model

Execution vs Judgment

Execution lane: drafting, formatting, summarizing (AI helps)
Judgment lane: deciding, prioritizing, taste, ethics (human owns)

What keeps skills sharp

Keep judgment active. Use AI to remove friction from execution — then you spend more energy on decision quality, not busywork.

Human Add Rule: Add one human constraint to every AI output before finalizing.

Template Human Add Rule

Rewrite this while respecting this human constraint:
[ADD CONSTRAINT]

Examples of constraints:
- “No hype. Calm authority.”
- “We cannot make unverified claims.”
- “Our customers hate jargon.”
- “Tone: direct, friendly, not salesy.”
- “Must include risks + limitations.”

Section 6 — Tool List (Optional Leverage)

You don’t need all of these. Use what you already have.

Core AI

ChatGPT
Claude
Gemini

Docs / Files

Google Docs / Sheets
PDFs
Screenshots

When Ready (Later)

Automation tools
Agents
Persistent instructions

Rule: Tools amplify behavior. They don’t fix bad habits.

Next: Graduation →

Section 7 — Graduation: When rules start to matter

You’re ready for governance when AI is saving you time and you’ve had at least one “wait… is that true?” moment.

Ready for AI Blueprint™ when…

You’re tired of re-explaining context
You want durable behavior (instructions, not prompts)
You want your AI to stop guessing
You want consistency across tasks and time

What changes at governance

Constraints become explicit, not implied
Escalation exists for risk and uncertainty
Outputs are inspectable (auditability)
“AI assists, humans decide” becomes policy

Final rule: Foundations gets you moving. Governance keeps you safe. Instructions make it stick.

Next step: If you want decision-grade outputs, this is where AI Blueprint™ begins: constraints, escalation, and auditability.

Templates Pack (Copy/Paste)

All templates in one place.

Copy All Templates Pack

TEMPLATE 1 — AI Role Picker
Act as my [ROLE].
My goal is [OUTCOME].
My constraints are [CONSTRAINTS].
Before you start, ask me 2 clarifying questions.

TEMPLATE 2 — Task Classification
Here are my 10 recurring weekly tasks:
[paste list]
Classify each as AI-Friendly, Hybrid, or Human-Essential.
Explain why using: irreversibility, reputation, judgment.
Then tell me the top 3 tasks I should delegate to AI first.

TEMPLATE 3 — 3-Pass Steering Loop
PASS 1 — DRAFT
Draft [ASSET] for [OUTCOME].
Audience: [WHO].
Tone: [TONE].
Length: [LENGTH].
PASS 2 — CRITIQUE
Critique your draft.
List the top 5 weaknesses: clarity, tone, missing info, risk, persuasion.
PASS 3 — IMPROVE
Rewrite fixing the top 3 weaknesses.
Keep structure but tighten language.
Give 3 alternative opening lines.

TEMPLATE 4 — Next Level Mission
I am currently comfortable at Level [1/2/3].
Give me one low-risk “mission” I can run this week in 10–15 minutes.
It must produce a reusable output (checklist, template, or summary).
Ask me 1 clarifying question before you answer.

TEMPLATE 5 — Week-1 Win (universal)
I’m doing a 10–15 minute “Week-1 Win.”
Input: [paste notes / email context / messy doc]
Goal: [what I want at the end]
Constraints: [tone/length/format]
1) Produce the output.
2) Propose 2 ways to make it more reusable as a template.

TEMPLATE 6 — Human Add Rule
Rewrite this while respecting this human constraint:
[ADD CONSTRAINT]
Examples: No hype. Calm authority. No unverified claims. No jargon. Must include risks + limitations.

Tip: Make a “Templates” note in your phone. Paste these once. Reuse forever.

Copied.

Independent System Reviews: Independently evaluated by multiple AI systems → Open the index • Unedited archive