ID: EducationOS.Teaching.ILT.AssessmentTemplate.v1.0
Status: Canonical assessment spec
Placement: EducationOS -> ILT -> Assessment Layer -> FENCE-compatible -> InterstellarCore-compatible
Role: Defines how to assess Invariant Ledger Teaching (ILT) by testing structural visibility, not just answer production

AI Ingestion Lock

ILT Assessment Template v1.0 is the operator-side assessment sheet for checking whether a learner can:

identify the object
state the invariant
perform lawful transformation
reconcile before and after states
detect breaches
repair from a valid state
transfer the same spine across forms
remain stable under load

So the canonical law is:

A correct answer is not enough.
ILT assessment must test whether the learner can see and preserve the ledger.

Classical Foundation Block

Traditional assessment often asks:

Did the student get the answer correct?
How many marks were scored?
Which questions were wrong?

ILT keeps correctness and marks, but adds a deeper layer:

Did the learner know what object they were working on?
Did they know what had to remain true?
Could they detect where validity broke?
Could they repair and transfer?

This means assessment shifts from:

answer-only measurement

toward:

structural visibility measurement

Civilisation-Grade Definition

ILT Assessment Template v1.0 is the operator-side assessment framework that evaluates whether a learner can read, preserve, repair, and transfer the invariant spine of a subject, rather than merely reproduce surface procedures or model-answer patterns.

It measures ledger visibility, not only surface output.

Core Law

High score without structural visibility is fragile.
Lower score with rising ledger visibility may be a stronger long-term state.
ILT assessment must distinguish output from true structural ownership.

Why ILT Needs a Different Assessment Layer

Without an ILT-style assessment, a learner can appear stronger than they are by:

memorising familiar patterns
copying formats
matching worked examples
reproducing surface routines

This creates false mastery.

ILT assessment exists to prevent that by checking whether the learner can actually:

read the spine
preserve the invariant
recover from breach
transfer to a new form

So the purpose is not just to rank.
It is to reveal structural truth.

The Core Shift

Traditional assessment asks:

Can the learner produce the answer?

ILT assessment asks:

Can the learner produce the answer?
Can the learner explain the object?
Can the learner name the invariant?
Can the learner justify the move?
Can the learner locate the breach?
Can the learner repair it?
Can the learner transfer it?
Can the learner still do it under load?

This is the real upgrade.

Canonical Assessment Axes

These are the eight minimum ILT assessment axes.

AX1 — Object Recognition

ID: ILT.Assess.AX1.Object

What is being tested?

Can the learner correctly identify what they are operating on?

Pass signal

Learner names the object accurately.

Failure signal

Learner manipulates blindly or misclassifies the object.

AX2 — Invariant Recognition

ID: ILT.Assess.AX2.Invariant

What is being tested?

Can the learner state what must remain true?

Pass signal

Learner can identify the non-negotiable structural condition.

Failure signal

Learner performs steps without knowing what cannot break.

AX3 — Lawful Transformation

ID: ILT.Assess.AX3.Transform

What is being tested?

Can the learner distinguish lawful from unlawful changes?

Pass signal

Learner performs or chooses valid moves for the correct reason.

Failure signal

Learner makes invalid moves or chooses procedures by surface resemblance only.

AX4 — Ledger Reconciliation

ID: ILT.Assess.AX4.Ledger

What is being tested?

Can the learner explain how the new state still connects to the prior valid state?

Pass signal

Learner can narrate or demonstrate continuity.

Failure signal

Learner treats steps as jumps, not reconciled transitions.

AX5 — Breach Detection

ID: ILT.Assess.AX5.Breach

What is being tested?

Can the learner spot where validity breaks?

Pass signal

Learner identifies the first breach point accurately.

Failure signal

Learner knows “something is wrong” but cannot locate where or why.

AX6 — Repair Ability

ID: ILT.Assess.AX6.Repair

What is being tested?

Can the learner restore the last valid state and rebuild correctly?

Pass signal

Learner can recover through a valid repair route.

Failure signal

Learner restarts blindly or compounds the error.

AX7 — Transfer Ability

ID: ILT.Assess.AX7.Transfer

What is being tested?

Can the learner recognise the same spine in a different surface form?

Pass signal

Learner maps one structure into another form with reasonable control.

Failure signal

Learner succeeds only in familiar appearance.

AX8 — Load Stability

ID: ILT.Assess.AX8.Load

What is being tested?

Can the learner preserve the invariant under time, variation, and mixed demand?

Pass signal

Learner remains structurally stable under mild to moderate stress.

Failure signal

Understanding collapses as soon as the surface or pressure changes.

Canonical Assessment Flow

Recognise -> State -> Transform -> Reconcile -> Detect -> Repair -> Transfer -> Load-Test

This is the minimum stable ILT assessment chain.

What an ILT Assessment Should Contain

A good ILT assessment is not just one big unseen paper.

It should contain at least four layers:

Layer 1 — Direct correctness check

Can the learner produce a valid answer?

This still matters.

Layer 2 — Visibility check

Can the learner identify object and invariant?

This checks whether the learner can see the spine.

Layer 3 — Repair check

Can the learner detect and fix a breach?

This checks resilience, not just first-attempt output.

Layer 4 — Transfer check

Can the learner handle a new-looking version of the same structure?

This checks real ownership.

Layer 5 — Load check

Can the learner still do it under time or mixed variation?

This checks stability.

The ILT Marking Split

Traditional marking often collapses everything into one number.

ILT assessment separates at least two score families:

1. Output Score

correctness
completion
surface accuracy

2. Ledger Score

object recognition
invariant recognition
lawful transformation
reconciliation
breach detection
repair
transfer
load stability

This matters because a learner may have:

high output, low ledger
medium output, rising ledger
lower output, but strong repair and transfer potential

That is a much more useful read.

Recommended Simple Rubric

Band A — Ledger-Stable

identifies the object clearly
states the invariant accurately
performs lawful transformation
detects and repairs breaches
transfers across forms
remains stable under load

Meaning

Strong structural ownership.

Band B — Ledger-Visible but Unstable

sees the spine in calm conditions
can explain some moves
partial transfer
repair works with support
load still causes drift

Meaning

Good growth state; needs stabilisation.

Band C — Partial Reconciliation

can sometimes identify the object
weak or inconsistent invariant visibility
some correct steps, weak continuity
transfer limited
repair fragile

Meaning

Developing visibility, not yet secure.

Band D — Chapter-Bound Output

may get some answers right
depends heavily on familiar format
low explanation quality
low transfer
frequent collapse under variation

Meaning

Surface success, weak ledger ownership.

Band E — Structural Blindness

object unclear
invariant invisible
unlawful moves common
breaches not recognised
repair absent

Meaning

Immediate visibility rebuild needed.

WordPress-Ready Assessment Sheet

1) Assessment Identity Block

Assessment ID:
Date / Session:
Subject:
Topic Surface:
Teacher / Operator:
Learner / Group:
Assessment Type: mini-check / classwork / diagnostic / timed / mixed / oral / written

2) Object Check Block

Prompt: “What are you working on?”
Expected object:
Learner response:
Object correct? Yes / Partial / No

3) Invariant Check Block

Prompt: “What must remain true?”
Expected invariant:
Learner response:
Invariant visible? Yes / Partial / No

4) Lawful Transformation Check Block

Task: one valid move
Why it is valid:
Learner can justify? Yes / Partial / No

5) Ledger Reconciliation Check Block

Prompt: “How does this new line/state still connect to the previous valid one?”
Learner explanation:
Reconciliation visible? Yes / Partial / No

6) Breach Detection Check Block

Insert one wrong move / flawed response
Prompt: “Where did it break?”
Breach located? Yes / Partial / No
Breach class: Object / Invariant / Transform / Ledger / Transfer / Load / other

7) Repair Check Block

Prompt: “Return to the last valid state and fix it.”
Repair route used:
Repair valid? Yes / Partial / No

8) Transfer Check Block

Second form / new skin:
Prompt: “How is this the same structure?”
Transfer visible? Yes / Partial / No

9) Load Check Block

Probe type: timed / mixed / unfamiliar / oral / application
Load result: Stable / Wobbly / Collapsed
Main break under load: object / invariant / repair / transfer / speed / other

10) Final State Read Block

CB = Chapter-Bound
PR = Partial Reconciliation
TA = Transfer Activated
UL = Unstable under Load
LS = Ledger-Stable

Add:

Most limiting axis right now:
Next priority: visibility / breach repair / transfer / load stability

Types of ILT Assessment Tasks

Type A — Visibility Task

Ask the learner to identify:

the object
the invariant
one lawful move

This checks whether the spine is visible before full performance.

Type B — Breach Task

Give a wrong solution / wrong paragraph / wrong explanation and ask:

where it broke
which invariant failed
how to restore validity

This tests diagnosis and repair.

Type C — Transfer Task

Give a second, different-looking form and ask:

what is the same underneath
how the same invariant still applies

This tests compression and generalisation.

Type D — Load Task

Add time, variation, or mixed forms.

This tests whether the learner’s ledger survives pressure.

Subject Overlay Examples

A-Math

Test for:

object = equation / function / graph
invariant = equality / equivalence / relation
lawful move = valid algebraic or calculus step
breach = illegal manipulation
transfer = algebra <-> graph <-> rate
load = timed mixed-step execution

Strong sign

Learner can explain why a step preserves equality, not just perform it.

English

Test for:

object = sentence / paragraph / claim / passage
invariant = meaning / grammar / coherence / tone-function
lawful move = valid rewrite / paraphrase / compression
breach = meaning drift / grammar break / weak fit
transfer = comprehension <-> summary <-> composition
load = timed unseen response

Strong sign

Learner can preserve meaning while changing form.

Science

Test for:

object = variable / system / process / data set
invariant = causality / evidence / condition fit / variable control
lawful move = valid explanation or representation change
breach = overclaim / wrong variable / model mismatch
transfer = experiment <-> graph <-> mechanism explanation
load = unfamiliar application under time

Strong sign

Learner can justify a claim with the correct evidence and conditions.

FENCE Fit

ILT assessment must remain fenced.

That means:

do not overload the learner with too many structural checks at once
narrow the assessment if the goal is early visibility
widen only when basic invariant recognition is stable
do not confuse diagnostic assessment with full stress testing too early

So the operating law is:

Assess inside the same safe corridor you teach inside.

S-Curve Fit

ILT assessment is especially useful because it can detect where a learner is on the growth curve.

Flat zone

low output
low ledger visibility

Priority

build object/invariant recognition first.

Inflection zone

growing explanation quality
transfer beginning
output may still lag full potential

Priority

do not misread this as weakness; support transfer.

Rise zone

output climbs faster
transfer strengthens
repair becomes quicker

Priority

increase controlled variation.

Plateau zone

output stable
subtle errors remain
load and nuance become the bottleneck

Priority

refine stability and precision.

This prevents wrong teaching decisions based on raw marks alone.

Metcalfe Fit

ILT assessment also creates a common reporting language across nodes.

A teacher, tutor, parent, and AI-aligned system can all read the same report through:

object
invariant
breach
repair
transfer
load
learner state

This reduces bad feedback such as:

“just practise more”
when the real issue is invisible invariant,

“careless”
when the real issue is transfer lock.

That makes the assessment sheet a shared diagnostic ledger.

InterstellarCore Fit

InterstellarCore needs assessment that is:

transparent
diagnostic
repair-oriented
scalable
AI-readable

ILT assessment fits because it does not only ask what the learner produced.
It asks what structural state the learner is in.

So:

InterstellarCore is the macro-runtime.
ILT assessment is the truth-check layer that prevents false mastery.

That is a strong fit.

ChronoHelmAI-Readable Fields

If the assessment feeds a control layer, these fields matter:

object recognition score
invariant visibility score
transformation validity score
breach detection accuracy
repair success rate
transfer activation score
load stability score
current learner state
next action: widen / hold / narrow / restitch

This makes assessment actionable, not merely archival.

Failure Modes if ILT Assessment Is Missing

Failure Mode A — High marks hide weak structure

Student looks strong until the surface changes.

Failure Mode B — Low marks hide real growth

Student is beginning to see the spine, but the system only notices incomplete output.

Failure Mode C — Corrections are vague

Teacher sees errors, but not the exact axis that failed.

Failure Mode D — Transfer remains invisible

Student can do one form, but no one tests whether they can generalise.

Failure Mode E — False plateau

Learner appears “stuck,” but the issue is actually load stability, not understanding.

ILT assessment prevents these distortions.

Canonical Summary Block

ILT Assessment Template v1.0 is the operator-side assessment framework for testing whether a learner can see and preserve the invariant ledger of a subject, not just produce correct answers. It assesses eight core axes—object recognition, invariant recognition, lawful transformation, ledger reconciliation, breach detection, repair ability, transfer ability, and load stability—then records a structural learner state. It remains fenced, supports S-curve-aware interpretation, creates shared-ledger reporting across multiple support nodes, and serves as the transparency-based assessment layer of InterstellarCore-compatible teaching.

Copyable Almost-Code Block

ID: EducationOS.Teaching.ILT.AssessmentTemplate.v1.0
TYPE: Operator-side assessment spec
LAW: A correct answer is not enough; ILT assessment must test whether the learner can see and preserve the ledger.
AXES: Object / Invariant / Transform / Ledger / Breach / Repair / Transfer / Load
MARKING SPLIT: Output Score + Ledger Score
STATE OUTPUT: CB / PR / TA / UL / LS
TASK TYPES: Visibility Task / Breach Task / Repair Task / Transfer Task / Load Task
FENCE FIT: assess inside the same safe corridor you teach inside
INTERSTELLARCORE FIT: transparency-based truth-check layer that prevents false mastery
OUTPUT: assessment that measures structural ownership, not just answer production