RC
rupertchesman.com
navigate   F facilitator notes
1 / 1
Facilitator notes

3-Hour Intensive · Updated June 2026

Claude AI
The 3-Hour Intensive

From chatbot to digital coworker in three hours — every Claude feature worth knowing, in the order you'll actually use them.

~3 hrs
Intensive
7
Modules
18
Lessons
1
Case study
Welcome

Built for business professionals, not developers

If you've used a chatbot once, you're ready. We start with how Claude actually works and walk all the way to a digital coworker and a governance brief.

Managers & operators

People who want Claude doing real work in their week.

Ops & finance teams

Drowning in spreadsheets, inboxes and admin.

Team leaders

Deciding how the team adopts AI — and defends it.

Anyone curious

No technical background needed.

"Claude is no longer a chatbot you ask things.
It's a family of models wrapped in products that do the work."

The old "type a prompt, copy the answer" model leaves 90% of the value on the table. Today is about moving from advisor to executor.

Agenda

The seven modules

01

The New Claude

Models, the heuristic, the product map.

02

Getting Great Output

RCTF prompting, Projects, Memory, Styles, Artifacts.

03

Claude That Does the Work

Cowork, Chrome, the Office add-ins, scheduled tasks.

04

Wiring Claude In

Connectors & MCP, Skills, Claude Design.

05

Trust & Governance

Data, certifications, the honest limits, a brief.

06

Your 30-Day Playbook

A staged plan you'll actually follow.

Bonus · Module 07 — Case Study: Doing an Australian company's books with Claude

01
Module One

The New Claude

How it works

A prediction engine — not a database, not a search engine

Brilliant at

Language work — drafting, summarising, reformatting, explaining, reasoning over the material you give it.

Where it guesses

Recalling things it was never given. That's hallucination — confident gap-filling, not lying. The fix: give it the source and verify what matters.

In 2026 Claude also uses tools — web search, code, your files, your apps — which removes much of the old weakness. The behaviour is the same: good input, clear job, check what counts.

The model family

Four tiers — cost & speed vs capability

Haiku 4.5

Fast & cheap. High-volume, low-latency work — classification, summarising, bulk jobs.

Sonnet 4.6

The workhorse. The default for most business work. Start here.

Opus 4.8

The flagship. Hard reasoning, long documents, multi-step agentic work.

Mythos (Fable 5)

Frontier. State-of-the-art — but currently suspended. Don't build on it alone.

The one rule

Start on Sonnet. Up to Opus when it's hard. Down to Haiku for speed and volume.

Model choice is gears. You don't drive everywhere in fifth — Sonnet for the commute, Opus for the hills, Haiku when you just need to move fast and cheap.

Cautionary tale

Fable 5: the most capable model in the world — for 72 hours

9 June 2026

Anthropic launches Fable 5, a Mythos-class model — state-of-the-art on most benchmarks.

12 June 2026

Suspended for all customers after a US export-control directive. Opus 4.8 stayed up.

Frontier-model access can be revoked overnight. Never depend on a single model — build with fallbacks.

Advisor → Executor

The product map you'll walk all course

Advisor

Claude chat

Think, draft, analyse, make Artifacts. The front door.

Executor

Cowork · Chrome · Office

Hand it whole jobs; it does the work while you supervise.

Fabric

Connectors · Skills · Memory

The connective tissue that makes it one system.

Every time you copy-paste between Claude and another app, that's a job an executor product should be doing for you.

02
Module Two

Getting Great Output

Prompting

RCTF — turn a vague ask into a reliable result

R · Role

Who Claude should be — sets the standard and perspective.

C · Context

Situation, audience, source material. 80% of quality lives here.

T · Task

One specific, clear instruction.

F · Format

Exactly what you want back — length, structure, tone.

"Make it good" isn't an instruction — it's a wish. The model is rarely the problem; the prompt is.

Make Claude yours

Stop sounding like AI. Stop re-explaining everything.

Projects

= your knowledge. Load documents and instructions once; every chat inside already knows them.

Memory

= your continuity. Claude remembers across conversations — view, edit, pause or reset it.

Styles

= your voice. Feed it your writing once; every draft starts sounding like you.

From answer to deliverable

Artifacts & file creation

Artifacts

Build and iterate real things inside Claude — "add a column", "make it blue". Co-editing, not re-generating. Live Artifacts even refresh their own data.

File creation

Download real .xlsx, .pptx, .docx, .pdf files — available to all users, including Free. "Turn this CSV into a model, a memo and a deck."

Hands-on

Rebuild your prompt with RCTF

Take the task you brought. Rewrite the prompt with all four of Role, Context, Task, Format. Run it, compare to your first attempt — then improve it with two follow-ups. Notice how much of the lift is just Context.

03
Module Three · the centrepiece

Claude That Does the Work

Cowork

"Cowork takes the outcome and handles the rest"

Regular chat

Can't touch your files. It advises; you do the multi-step work.

Cowork

Give it a goal and a folder; it reads, edits and creates files and returns finished deliverables.

Human oversight by design: it shows a plan and waits for approval, and loops you in before anything significant. That gate is your safety belt.

Acting where you work

Claude in Chrome & the Office add-ins

Office add-ins

Claude inside Excel, PowerPoint & Word (Outlook in beta). Edits are highlighted for review; nothing goes out until you click send.

Chrome — handle with care

A browser agent that acts on your tab. Prompt injection is real (better, not zero). Keep it away from banking, passwords and sensitive data.

Proactive Claude

Scheduled tasks — work that happens without you

Recurring

Hourly, daily or weekly — a morning brief, a weekly report. Each run is a fresh session.

The limit

Cowork tasks run only while the computer is awake and the app is open.

The rule

Keep outputs as drafts; watch the first runs; earn autonomy run by run.

Demo + do

Delegate one real, multi-step job

If you have the desktop app, point Cowork at a folder of low-stakes files and give it a genuine job. Approve its plan, watch it work, correct it once. Notice it asks permission before anything that matters.

04
Module Four

Wiring Claude In

The connective tissue

From features to a system

Connectors & MCP

Plug Claude into Gmail, Drive, Slack, Asana and dozens more. MCP is "a USB-C port for AI" — why the list keeps growing. Review every permission.

Skills

Graduate your best prompt into a packaged, shareable capability that runs your way every time — install one or build with the Skill Creator.

Claude Design

Describe a deck, one-pager or landing page and refine it live on a canvas. The fastest path from idea to something to show.

Skill = the moves · Project = the knowledge · Connectors = the reach · Cowork = the execution.

05
Module Five

Trust & Governance

Capability is half the decision. The other half is "can I defend this to risk, legal and the board?"

The buy-in answers

The three questions every business asks

Is our data used to train it?

No — not on business plans by default. Team, Enterprise and the API are excluded from training. For sensitive work, use that login.

Is it certified?

SOC 2 Type I & II, ISO 27001:2022, ISO/IEC 42001:2023; GDPR via DPA + SCCs; HIPAA BAA on Enterprise.

Is it safe?

Ad-free pledge, the Responsible Scaling Policy (active ASL-3), and Constitutional AI — transparent and auditable.

Honest limits

Name the risks — then hand IT a one-pager

The limits, plainly

Hallucination (verify what matters) · prompt injection · human-in-the-loop on autonomous runs · vendor risk (keep fallbacks) · benchmark claims are vendor claims.

The Governance Brief

A one-page summary — plan, data stance, certifications, the guardrails you'll mandate — that turns "I tried a cool AI thing" into "here's a governed plan."

Teaching the limits isn't hedging — it's how you earn the right to recommend the tool.

06
Module Six

Your 30-Day Playbook

No new features — just synthesis and commitment. You'll remember 30% of the demos, but keep 100% of the plan you write.

The staged plan

Four weeks, advisor to executor

Week 1

Get fluent

One task daily; Sonnet default; build your prompt library; turn on Memory.

Week 2

Make it yours

A custom Style; your first Project; file creation on the doc you dread.

Week 3

Delegate

A real Cowork job; the Office add-ins or a scheduled task; watch the first runs.

Week 4

Make it permanent

Turn a task into a Skill; finish the Governance Brief; pick your next course.

Plan: start on Pro ($20) → Max if you hit limits weekly → Team for groups → Enterprise for SSO/audit/keys/BAA.

07
Bonus · Case Study

Doing the Books with Claude

A true, end-to-end account of running a GST-registered Australian company's books — Claude Cowork orchestrating, Claude in Chrome driving QuickBooks Online. Every figure is real.

The setup

Three tools, working together, supervised

Claude in Chrome

Drives the QuickBooks UI directly — no API needed.

Cowork + Python

Orchestrates the multi-step job; a sandbox does the heavy data work.

Gmail connector

Pulls receipts and remittance advices to support the work.

The golden rule set up front: "Prior lodged years — do not touch." Watching Claude respect it is the whole lesson.

What we achieved

The numbers at a glance

Mystery customer payment resolved$13,183.50
Main bank accountReconciled, $0 diff
Broken bank rules cleaned13 (0 left)
Prior-period GST surfaced for review~$6,940
Accountant workload5 specific questions
The real lesson

The best move an AI made all day was to stop.

Three times — the Wise USD opening balance, the prior-year journals, and the BAS — Claude hit something that touched lodged returns or needed professional judgement, and handed back instead of guessing.

AI does the 90% that's tedious, surfaces the 10% that needs a human, and leaves a clean trail for both. Claude is not a registered tax agent — and that's the point.

You didn't just learn about Claude

You walked in with a chatbot. You're leaving with a digital coworker — and a plan.

The people who get value aren't the most impressed — they're the ones who delegate one real task next week. Pick it now, and open the playbook Monday.

Next steps

Keep going

Open the dashboard

18 interactive lessons to revisit, at your own pace, any time.

Go deeper

Mastering AI Tools for hands-on depth, or AI Agents for the frontier.

Bring it to your team

A private session on one of your real workflows — corporate training.

rupertchesman.com

Rupert Chesman — AI educator, filmmaker, author of The AI-Native Playbook · ru@rupertchesman.com