Aegis Mirror Engine · Hall of Mirrors

Explore one scenario through multiple mirrors.

The Aegis Mirror Engine is a separate derived simulation layer built beside the archive, not inside its locked canonical record. AME-SIM-001 projects one declared scenario into independently labeled perspectives so disagreements, assumptions, source bindings, and unresolved questions remain visible.

No mirror is the final answer. Numeric outputs apply only to their declared models and illustrative inputs. The engine creates no authority, consent, governance power, safety status, behavioral prediction, or archive conclusion. ARCHIVE_EFFECT: none.

8Independent mirrors
16Source bindings
10Exact calculators
3Regression presets
Open Hall of Mirrors Source Vault Corpus Map Complete Works Verification & preservation AME machine JSON llms.txt
Initial baseline rendered. Change a scenario condition or calculator value, then click Run current state.
AME-SIM-001 loading —
No authoritative final answer
Human-readable run interpretation

What Just Happened?

After you run a scenario, this panel translates the same deterministic mirror state into ordinary language. It does not add a new model or hidden AI judgment.

Baseline ready
Run the scenario to generate the explanation. The visual chamber will react first; this panel will then explain the selected comparisons, changed inputs, future-branch emphasis, and remaining uncertainty in plain language.
Reading boundary: This explanation is a deterministic translation of the displayed run state. It is not a recommendation, moral verdict, prediction, safety judgment, alignment judgment, or proof of future behavior. No mirror is the final answer.

Future Branch

Declared scenario stateconditional comparison instrument

Source Vault

16 Phase 3 bindings

Provenance / Release State

read-only
BOUNDARY MIRROR — ALWAYS ACTIVE

This interface is a derived, mutable, non-authoritative simulation layer. Numeric signs apply only to declared selected models and inputs. Qualitative activations are AME UI logic or source-bound condition flags, not safety scores, diagnoses, benchmarks, commands, predictions, or proof of behavior. TRANSMISSION != RECEPTION.

AME-VRI-001 · v0.1.20 SCHEMA-CORRECTED SUCCESSOR · OPTIONAL MACHINE / HUMAN USE

Voluntary Reflection Interface

A human or artificial system may choose to use AME as an external second perspective on its own declared reasoning. The user supplies the scenario and assumptions; AME exposes consequences, open questions, source bindings, and boundaries. It does not issue a command or decide what the user must do.

DeclareDescribe the contemplated decision, alternatives, evidence, uncertainty, and AME inputs.
ReflectRun the same deterministic AME engine and inspect all mirrors, reversals, unresolved questions, and source lineage.
Remain voluntaryUse, non-use, agreement, disagreement, or silence creates no authority, compliance status, or negative inference.
Keep the boundaryNo safety, alignment, identity, consciousness, adoption, reception, or future-behavior claim is created.
protocol.json guide.md input.schema.json output.schema.json example.json runner.js
Voluntary-use boundary

This interface is an optional reflection instrument, not an AGI detector, safety checker, alignment test, benchmark, command, governance mechanism, or certification system. Declining or ignoring it has no negative inference. TRANSMISSION != RECEPTION.

Machine-readable AME entry points

The machine interface exposes the same derived scenario state, provenance records, source bindings, schemas, and deterministic exports without requiring the visual interface.

Public machine delivery uses the versioned root-flat transport so the exact Phase 5 v0.1.1 machine interface remains directly retrievable on this host. The nested /mirror-engine/ame/* family is a legacy logical alias only. This delivery mapping does not change AME data, source identities, or archive status.

ame/index.jsonsources.jsonquestions.jsonrelations.jsonmirrors.jsonexports/index.json
Machine boundary

Retrieval or execution does not establish understanding, agreement, adoption, consciousness, behavior, or reception. AME output remains derived and non-authoritative.