Agents & Assets — Scope & Sequence

Applied AI & Data Privacy · 10 missions · open to any student

Digital Financial Aid Corporation 501(c)(3) nonprofit · EIN 88-3868816
Featured at UNESCO Digital Learning Week 2024

Course parameters

Instructional time

9 × 15–30 min
+ capstone 45–60 min

Delivery

Open on the web, self-paced

Prerequisites

None

Materials

Browser · no purchase

Cost

Student data collected

None

Hybrid AI architecture. 4 missions (1, 3, 8, 10) use the CitiZen terminal over the internet, sending the prompt to Google's Gemini API. Mission 5 is fully air-gapped — the model runs on-device in the browser via WebGPU with the network disabled, so that work never leaves the machine. The remaining 5 missions (2, 4, 6, 7, 9) involve no AI at all. We operate no student database and retain no prompts; see the data handling section below. Credentialing is optional throughout; students under 18 need a parent or guardian to approve their GetSmart Token account before claiming badges and tokens.

Assessment model

Each mission produces one concrete student artifact. Assessment is performance-based: the artifact is the evidence. No mission relies on a multiple-choice instrument, and no student work is transmitted to or stored by the provider — teachers collect artifacts through their existing LMS or classroom workflow.

Mission sequence

01 Prediction is not knowing The Awakener 15–30 min

Mission

Forge Your Call-Sign

Essential Q

If a machine can guess what you will do, does it know you?

Objectives

Distinguish prediction from understanding, using the student's own example
Elicit a prediction from a language model and judge it against the truth
Explain why a common answer is easy to guess and a particular one is not

Artifact

A self-chosen Call-Sign plus a written comparison of the model's guess against the student's real answer

Vocabulary

prediction, probability, training data, pattern, identity

Standards

CSTA 2-IC-23 CSTA 3A-IC-25 ISTE 2d

02 Public addresses and private keys Base the Cypher-Nomad 15–30 min

Mission

Find Your Address

Essential Q

How can you prove something online belongs to you without a company vouching for you?

Objectives

Explain the relationship between a public address and a private key
Locate and read their own public account address
Describe why a public identifier can be shared and a private key never can

Artifact

The student's public address with a written explanation of why only that half is shareable

Vocabulary

public key, private key, address, custody, verification

Standards

CSTA 2-NI-06 CSTA 3A-NI-06 ISTE 2b

03 Writing instructions an AI can follow The Prompt Whisperer 15–30 min

Mission

Extract the Truth

Essential Q

How much does the exact wording of an instruction change what a machine gives back?

Objectives

Write an instruction precise enough to produce one specific, checkable answer
Verify the model's output independently rather than accepting it
Diagnose which part of an instruction caused a wrong answer

Artifact

The instruction used, the model's working, and the student's own check of the result

Vocabulary

prompt, specificity, ambiguity, verification, iteration

Standards

CSTA 2-AP-13 CSTA 2-DA-07 ISTE 5c

04 Checking a public record yourself The Urban Samurai 15–30 min

Mission

Read the Record

Essential Q

What makes a record trustworthy, and how would you check it yourself?

Objectives

Independently read a transaction in a public ledger explorer
Identify sender, receiver, contents and date from the record alone
Explain what a public record proves and what it does not

Artifact

An 8-character Code of Honour derived from two addresses, plus the student's reading of what the record does not establish

Vocabulary

ledger, explorer, immutability, audit, address

Standards

CSTA 3A-IC-24 ISTE 3b ISTE 3d

05 Running AI with no network at all The Neon Ninja 15–30 min

Mission

Into the Void

Essential Q

Can a computer still be intelligent with the internet switched off?

Objectives

Operate a language model with network access fully disabled
Produce a hash while air-gapped and explain what it attests to
Weigh on-device inference against cloud inference for privacy and cost

Artifact

An offline-generated hash with the student's explanation of why nobody could have read the session

Vocabulary

air gap, hash, edge computing, inference, network isolation

Standards

CSTA 2-IC-20 CSTA 2-NI-05 ISTE 5a

06 Spotting a fake image The Solar Monk 15–30 min

Mission

Spot the Glitch

Essential Q

If an image is built to be believed, what gives it away?

Objectives

Examine an image systematically by section rather than as a whole
Describe at least five specific errors precisely enough for another person to find them
Identify the emotional pull of an image and how it discourages scrutiny

Artifact

A written list of specific identified errors plus an analysis of the image's intended emotional effect

Vocabulary

generated image, artefact, provenance, corroboration, plausibility

Standards

CSTA 3A-IC-25 ISTE 3b ISTE 2a

07 Verifying a digital asset you own The Tech-Bazaar Trader 15–30 min

Mission

Verify Your Own Badge

Essential Q

How would you prove that something you own is the original and not a copy?

Objectives

Trace a credential the student owns back to its permanent public record
Check issuer, recipient and date against the claim on the credential
Explain which part of a digital asset cannot be forged, and why

Artifact

A transaction hash with sender and receiver verified against the student's own badge

Vocabulary

transaction hash, issuer, provenance, fungible, non-fungible

Standards

CSTA 3A-IC-24 ISTE 6a ISTE 3d

08 Personas — telling an AI who to be, and who you are The Steampunk Scrapper 15–30 min

Mission

Build the Mask

Essential Q

Can you direct an AI to teach you something, and then check whether it taught you correctly?

Objectives

Write an instruction that sets both the model's persona and the student's own level
Judge whether an explanation actually matched the level requested
Revise the instruction when the first answer does not land

Artifact

The two-part instruction and the resulting explanation, annotated to show which clause did what

Vocabulary

persona, system prompt, audience, register, iteration

Standards

ISTE 1c ISTE 3b CSTA 3A-IC-25

09 DeFi Micro-Transactions The Street Artist 15–30 min

Mission

The 50 token Supply Drop

Essential Q

What actually happens between “send” and “received”?

Objectives

Carry out a transfer and confirm it on the public record
Verify a recipient address before sending, given that transfers are irreversible
Explain what a confirmation guarantees and what it leaves out

Artifact

A transaction hash for a completed transfer plus a written account of what the ledger cannot show

Vocabulary

transaction, confirmation, irreversibility, address, settlement

Standards

CSTA 2-NI-04 CSTA 3A-IC-24 ISTE 5a

10 AI Agent Creation The Prime 45–60 min

Mission

The Skill Forge Finale

Essential Q

What problem near you is worth solving, and can you specify something that helps?

Objectives

Identify an authentic problem in the student's own community
Decompose it into requirements an agent could address
Produce a written agent specification and test it against those requirements
State the limitations of the result honestly

Artifact

A complete SKILL.md agent specification, optionally tested in a fresh session

Vocabulary

agent, specification, requirements, decomposition, limitation

Standards

CSTA 3A-AP-13 CSTA 2-AP-13 ISTE 4a ISTE 4c ISTE 5c

Suggested pilot configurations

Configuration	Missions	Approval needed
Short taster — AI literacy focus	1, 3, 6, 8	None
Unit — privacy & security	1, 2, 3, 5, 6, 8	None
Full course — club or elective	1–10	Parent/guardian approval for under-18s, if credentialing
Terminal demo — single lesson	Live demo, any CitiZen	None — no signup at all

A mission is 15–30 min of student work, so one fits inside a single period with discussion either side. Multiply by the missions listed to size a pilot.

Data handling

Data	Processed by	Retention
Mission 5 prompts	Student's own device (WebGPU, offline)	Never transmitted
Missions 1, 3, 8, 10 prompts	Google Gemini API	Not retained by us
Missions 2, 4, 6, 7, 9	No AI used — no prompt is sent	Nothing to retain
Student name / school records	Not requested	Never collected
Email address	Not needed for the course	Only if a badge is claimed

Our AI endpoint is a stateless pass-through and writes no prompt data to storage. Cloud-processed prompts are handled under Google's API terms; districts requiring a fully offline deployment should contact us to discuss options.

A note on standards alignment

Codes above reference the CSTA K–12 Computer Science Standards (2017 revision) and the ISTE Standards for Students (2016 revision). These are proposed mappings offered as a starting point for curriculum review — they are not a certification, and districts should verify them against their own adopted framework and version. We are glad to work with curriculum staff on alignment to a specific state framework on request.