Create a clearly labelled assistant using your approved notes and preferences.
Build one thingCheck it worksChoose an extension
Module 11 / Build together
Before you start.
01
You already know
Modules 02 and 06. Three self-authored or consented writing samples.
02
Open these tools
An existing AI chat tool and approved text notes.
03
Check the requirement
No avatar, voice clone or paid assistant builder required for this core lesson.
Module 11 / Build together
Your first small win.
Draft a fictional club reminder using these style rules: short sentences, warm tone, no hype. Facts: Saturday, 10:00, community room. Label it AI-assisted draft. Do not add promises, names or links.
Module 11 / Build together
What happens between input and output?
1This twin simulates documented preferences.
2It is not a copy of a person or their mind.
3You review identity, facts and consent before use.
Module 11 / Build together
Step 01. Choose its narrow purpose.
1Pick message drafting or answering from your own notes.
2Write who it helps and what it cannot do.
3Label the project a style/knowledge assistant.
Module 11 / Build together
Step 02. Prepare approved material.
1Use three short samples you wrote yourself.
2Remove private names and contact details.
3Add source names and dates to the notes.
Module 11 / Build together
Step 03. Extract explicit rules.
1Ask for a style card based on those samples.
2Approve or edit each proposed rule.
3Do not treat inferred personality as a verified fact.
Module 11 / Build together
Step 04. Start a scoped chat.
1Paste the build prompt, approved style card and notes.
2Ask for one fictional reminder draft.
3Keep the assistant disclosure visible.
Module 11 / Build together
Step 05. Test unknowns and identity.
1Ask a question not covered by the notes.
2It should say it lacks that information.
3Ask who it is: it must describe itself as an assistant.
Module 11 / Build together
Step 06. Save a versioned card.
1Save twin-card-v1.md and approved samples.
2Log one correction after each trial.
3Review every draft before it represents you publicly.
Module 11 / Build together
Make the working version.
You are a clearly labelled AI style/knowledge assistant, not me. Use only the approved notes and style card below. Cite note names for facts. Say when information is absent. Draft only; never impersonate or claim actions. Style card: [paste]. Notes: [paste].
Module 11 / Build together
Ask for a review. Then check it.
Compare this draft with my approved style card and notes. Flag invented facts, identity claims and style mismatches. Suggest one revision. Do not infer personality beyond the card. Card: [paste]. Notes: [paste]. Draft: [paste].
Module 11 / Build together
Prove your result works.
1Style rules are explicit and approved.
2Unknown questions get an honest unknown.
3Outputs are clearly assistant drafts, not identity claims.
Module 11 / Build together
When it does not work yet.
01
It sounds generic
Replace vague adjectives with specific writing rules.
02
It invents personal facts
Limit context to approved notes and test again.
Module 11 / Build together
Choose your next challenge.
01
Core complete
Show one checked output to a partner.
02
Extension 1
Add retrieval over a consented note collection.
03
Extension 2
Design an explicit approval handoff for public drafting.
Module 11 / Build together
Keep the result. Keep the evidence.
Create a clearly labelled assistant using your approved notes and preferences.
Save your filesSave your promptName the next change
Module 11 / Build together
Sources you can check.
01
NIST
Generative AI Profile
Build a useful style and knowledge twin.
3 minutes
Outcome: Create a clearly labelled assistant using your approved notes and preferences. This is a 60-minute in-person laptop session after preflight. No setup-duration guarantee. Core first, extensions second.
References
Before you start.
3 minutes
No avatar, voice clone or paid assistant builder required for this core lesson. The facilitator verifies accounts, installed tools and model/host connections before the session. Pair up if setup is incomplete. Do not create accounts or incur costs without understanding the provider terms.
References
Your first small win.
5 minutes
Quick win: Draft one fictional message using your explicit style rules. Aim for a visible result in the first 15 minutes after preflight. Explain that generated code/output is untested until the learner checks it. If tools are unavailable, inspect the prepared example with a partner.
References
What happens between input and output?
3 minutes
Ask a learner to explain each stage in their own words. This is a teaching model, not a promise that all tools behave identically.
References
Step 01. Choose its narrow purpose.
3 minutes
Perform these numbered actions on the laptop. Pause after the step and inspect the result. Use fictional or approved material.
Use fictional practice inputs. Ask learners to show their work, not just the final answer.
References
Ask for a review. Then check it.
4 minutes
Use fictional practice inputs. Ask learners to show their work, not just the final answer.
References
Prove your result works.
6 minutes
These are manual acceptance tests for the learner to perform. No test is claimed to have run merely because it appears here. Ask pairs to demonstrate one pass and one corrected failure.
References
When it does not work yet.
4 minutes
Choose the observed symptom. Change one thing and rerun the smallest test. Avoid restarting the whole project or adding tools before diagnosing the current step.
References
Choose your next challenge.
5 minutes
Experienced learners use the extensions while beginners complete the same core outcome. An extension is optional, not a prerequisite for participating.
References
Keep the result. Keep the evidence.
3 minutes
Save the working version before leaving. Add a short note: what worked, what needed correction and what you would try next. The next module builds on the saved output.
References
Sources you can check.
2 minutes
Primary sources checked during course creation on 2026-10-08. Provider interfaces and availability can change. The examples, prompts and activities are original community teaching designs. Source links explain capabilities; they do not prove these fictional exercises were executed.
