Reference Implementation for Source-Bound Runtime Reconstruction and Evidence-Governed Computation

SubstrateX Aperture™ is the browser-local reference implementation of SubstrateX Runtime Evidence Infrastructure. Created by Arjay Asadi, Aperture is designed to preserve evidentiary authority continuously—from qualified source records through canonical reconstruction, behavioral measurement, instrument projection, human investigation, Passport formation, export, and preservation.

Its defining engineering property is authority continuity:

Every authoritative measurement, marker, classification, finding, interpretation, visualization, and export must remain accountable to the same source-bound Current Evidence Run.

Aperture is not a collection of dashboards assembled around a log analyzer. It is an evidence-governed computational architecture in which operational records are qualified, assigned canonical identity, reconstructed as role-bearing runtime trajectories, measured through versioned contracts, and exposed through bounded instruments and investigations that cannot create independent evidentiary authority.

The architecture is engineered around a simple constraint:

No downstream surface may become more authoritative than the evidence, method, temporal horizon, and status from which it was produced.

This page describes how that constraint is implemented.

From Research Instrument to Reference Architecture

Aperture began as an experimental environment for making long-horizon computational behavior visible in recorded interactions.

As the implementation developed, the engineering problem became larger than trajectory calculation or regime visualization.

An investigator also needed to know:

• which source records produced the reconstruction;

• how those records were qualified;

• how events, roles, timestamps, and dependencies were canonicalized;

• which values were source-observed, normalized, derived, heuristic, or projected;

• which evidence was eligible at each temporal position;

• when a marker became available;

• which method and version produced a measurement;

• whether different interfaces referred to the same runtime;

• what an instrument was permitted to present;

• what an interpretation was permitted to claim;

• how missing or conflicting evidence was represented;

• and what would remain inspectable after the browser session ended.

These requirements transformed Aperture from a research instrument into a reference architecture for Evidence-Governed Runtime Reconstruction.

A worldline cannot select its own source. An instrument cannot invent telemetry because its display requires a value. An operational lens cannot silently change a threshold. An interpretation cannot convert a proxy into a source fact. A Passport cannot grant authority absent from the evidence. An export cannot retain a conclusion while discarding the source and method lineage supporting it.

The architecture exists to enforce those separations while preserving the relationships among them.

The Authority Path

The logical architecture follows one controlled path:

Source records → Source qualification → Canonical runtime → Versioned computation → Current Evidence Run → Bounded projections → Guided Investigation → Certified Runtime Evidence Record → Runtime Evidence Passport → Evidence Commons preservation

Each transition has an explicit boundary.

• Source resolution does not perform diagnosis.

• Qualification does not establish failure.

• Canonicalization does not validate a scientific construct.

• Measurement does not establish cause.

• Marker computation does not create an external outcome.

• An instrument finding does not modify canonical telemetry.

• An interpretation does not become a source observation through review.

• Guided Investigation does not rewrite the run.

• Certification establishes conformance to the declared evidence-formation contract; it does not certify truth.

• Passport formation identifies the issued evidence record; it does not strengthen its claims.

• Export does not strengthen a claim.

• Preservation does not increase evidentiary authority.

The Current Evidence Run is the authoritative computational object connecting the active stages. When issuance requirements are satisfied, its eligible evidence is sealed as a Certified Runtime Evidence Record. The certified record carries the run's content and lineage forward; it does not become a second, stronger version of the evidence.

Governing Engineering Invariants

Aperture is organized around a stable set of engineering invariants.

Source primacy

Every authoritative result must resolve to identifiable source material or to an explicitly disclosed external dependency. Derived values must preserve the observations and transformations on which they depend.

Canonical identity

One qualified source set and computation profile must resolve to one identifiable Current Evidence Run. Routes, instruments, visualizations, investigation chapters, and exports cannot create competing run identities.

Evidence before projection

The canonical runtime and governed evidence objects are formed before visualization, narrative interpretation, operational translation, or export formatting.

Authority non-amplification

No projection may support a stronger claim than its source coverage, method, temporal horizon, validation status, uncertainty, and authority class permit.

Instrument non-sovereignty

Instruments read authorized evidence and registered models. They cannot create source facts, rewrite canonical telemetry, move formal markers, change Passport identity, or establish an independent history of the runtime.

Temporal admissibility

Every temporal claim must identify its coordinate, ordering relation, eligible evidence horizon, marker authority, and prospective or retrospective status.

Explicit absence

Missing, unavailable, ambiguous, conflicting, not-computable, candidate, unconfirmed, and not-observed states remain distinct from zero, normality, or a positive finding.

Contextual non-interference

Operational Worlds may change language, guidance, examples, and investigative emphasis. They do not silently change canonical telemetry, marker logic, regime rules, or evidence eligibility.

Preservation closure

A preserved conclusion must travel with the source identity, transformation lineage, method version, status, limitations, uncertainty, and claim boundary required to examine it.

These invariants are not interface conventions. They are architectural constraints that must remain true across ingestion, computation, investigation, replay, export, and preservation.

Source Provider and Ingestion Architecture

Aperture begins with operational material already available to the operator, including:

• transcripts;

• logs and distributed traces;

• JSON, JSONL, CSV, and structured event records;

• tool and agent traces;

• incident communications;

• workflow records;

• tickets and change histories;

• guided samples;

• and registered validation bundles.

The source-provider architecture separates responsibilities that ordinary upload interfaces often collapse:

1. resolving the selected source;

2. preserving source identity and, where available, original bytes or stable references;

3. recording available integrity information;

4. identifying source type, adapter, and fixture metadata;

5. qualifying the material for supported reconstruction and measurement;

6. recording disclosure and verification limits;

7. retaining it under the applicable browser-session policy; and

8. exposing explicit failure states when the source cannot be resolved or qualified.

Successful parsing is not equivalent to evidentiary sufficiency. A file may be syntactically valid while lacking timestamps, roles, outcome markers, tool state, or other information required for a proposed measurement.

Source content remains data.

Prompt-like instructions, malicious text, scripts, markup, or executable fragments contained inside a record do not acquire authority to alter Aperture policies, contracts, registries, computation, or exports. The architecture treats imported material as untrusted evidence input, not as operational instruction.

Qualification Before Computation

Qualification determines what the available source can support before the complete analytical stack is invoked.

The qualification layer may assess:

• source identity and origin;

• schema and adapter compatibility;

• event and turn coverage;

• temporal coverage and timestamp quality;

• actor and role disclosure;

• tool and workflow representation;

• model and configuration disclosure;

• supplied outcome and marker availability;

• missing, duplicated, malformed, or conflicting fields;

• transformation history;

• integrity information;

• redaction and disclosure state;

• and constraints on independent verification.

Qualification produces support states rather than a binary declaration that the source is “valid.” One record may support event-order reconstruction but not clock-time duration. Another may support role analysis but lack an independent outcome marker. A third may support replay while withholding source content from external verification.

The resulting limitations become part of the Current Evidence Run. They cannot be suppressed because a downstream visualization expects a complete value.

Role-Aware Canonicalization

Aperture reconstructs computational behavior as an interaction among participants rather than flattening every event into one undifferentiated stream.

Canonical events can preserve:

• source identity and source span;

• original actor and role labels;

• normalized role classes;

• role-assignment provenance;

• confidence, ambiguity, and competing assignments;

• operational authority relationships;

• role transitions and handoffs;

• objective and constraint ownership;

• tool and service participation;

• temporal fields and dependencies;

• content and event type;

• and transformation history.

This structure supports investigation of cross-role displacement, authority conflict, handoff continuity, role-relative lag, tool-loop pressure, unintegrated correction, and observer-state divergence.

Role topology does not infer private identity, intent, motive, blame, negligence, or legal responsibility. It represents operational relationships supported by the record.

Canonicalization also preserves the distinction between a role declared in the source, a role inferred through a registered adapter, and a role assigned by an operator. Those classes may inform the same investigation, but they do not carry the same evidentiary authority.

Essential-First Evidence Formation

A browser-local system should not leave the entire investigation without a stable evidence object because a large appendix, visualization, or export model is still being constructed.

Aperture therefore uses an essential-first evidence commit.

The minimum evidence-bearing core is committed when primary reconstruction and governed computation complete. It may include:

• runtime and source identity;

• canonical events and frames;

• the runtime spine;

• role and interaction context;

• registered measurements and availability states;

• temporal coordinates and markers;

• regime history;

• method and configuration versions;

• core findings;

• and claim boundaries.

Large reports, visual models, supplemental projections, appendices, and export structures can be hydrated afterward without changing the identity of the governed core.

The lifecycle progresses through five evidence states:

Acquired → Qualified → Computed → Issued → Preserved

An animation, completed chart, route change, or interface transition cannot advance this lifecycle. Evidence state is controlled by the required computational objects and contracts.

If a later stage fails, the system should preserve the strongest complete state already formed rather than representing the entire investigation as either fully successful or nonexistent.

The Current Evidence Run and the Certified Runtime Evidence Record

The Current Evidence Run is the sole run-wide computational authority while an Aperture investigation is active.

It binds the objects required by the observatory, including:

• qualified source identity and coverage;

• canonical events and runtime frames;

• the runtime spine and playback sequence;

• actors, roles, tools, and dependencies;

• temporal coordinates;

• registered measurements and signals;

• regime and marker authority;

• instrument models and findings;

• availability, uncertainty, and conflict states;

• human interpretations and review status;

• Passport data;

• and export and preservation context.

The Current Evidence Run is not a claim that the runtime has been completely observed or scientifically validated. It is the authoritative object representing what Aperture computed from the qualified source under a declared method and configuration.

When the declared issuance requirements are satisfied, the eligible evidence in the Current Evidence Run is sealed as a Certified Runtime Evidence Record (CER). The CER is the source-bound, versioned evidence artifact that carries the canonical runtime, methods, measurements, findings, missingness, provenance, review state, replay conditions, and claim boundaries into Passport issuance, export, and preservation.

Certification establishes conformance to the declared record-formation contract, including identity, required provenance, integrity state, method and version lineage, availability, claim boundaries, and preservation readiness. It does not establish source authenticity, completeness beyond the supplied record, objective truth, universal scientific validity, causal correctness, safety, legal admissibility, or deployment approval.

The Current Evidence Run and the CER are therefore two lifecycle states of one continuous evidence authority: active computation followed by an issued, preservation-ready record. Certification does not create a competing runtime or amplify the epistemic authority of its contents.

The Canonical Runtime Spine

The canonical runtime spine is the shared coordinate, reconstruction, and replay structure underlying Aperture.

Each position may contain:

• canonical frame and source coordinates;

• active actors and role relationships;

• source-linked content and events;

• clock, turn, event, dependency, frame, and symbolic-time coordinates;

• continuity, displacement, recurrence, pressure, and role measurements;

• regime state;

• boundary, outcome, and recovery markers;

• evidence and availability status;

• applicable uncertainty and conflict;

• method and version identifiers;

• and direct source references.

The worldline is projected from this spine.

It is not an arbitrary curve drawn through selected scores. Every visible position must resolve to a canonical frame, registered measurements, applicable events, source material, and claim boundary.

This makes aggregate structure traceable to local evidence. A visible deformation can be inspected at its source position. A regime transition can be traced to the frames satisfying its entry and persistence rules. Replay becomes traversal of the evidence object rather than animation over detached chart data.

Unified Temporal Authority

Aperture resolves temporal meanings through one shared temporal authority.

This prevents Seismo, Chronos, Runtime Reconstruction, Guided Investigation, operator briefs, and exports from independently deciding when weakening, boundary formation, Basin Exit, observable failure, or recovery occurred.

The architecture distinguishes:

• t_aw — earliest supported weakening or awareness marker under a declared method;

• t_candidate — first qualifying candidate-boundary marker;

• t* — confirmed Basin Exit under the registered boundary and persistence contract;

• t_f — independently supplied or qualified observable failure marker; and

• t_r — sustained Recovery or re-entry marker under declared criteria.

These markers are not interchangeable.

A candidate boundary is not a confirmed transition. A confirmed Basin Exit is not automatically an observed failure. A source-supplied outcome is not automatically a scientifically validated ground truth. Re-entry does not establish sustained Recovery.

Formal lead time is available only when t* and t_f are both present, qualified, ordered, and expressed on compatible coordinates:

Δt = t_f − t*

Without an observable failure marker, Aperture may report a warning interval, candidate interval, confirmed transition, or open post-exit observation window. It cannot manufacture formal lead time.

Retrospective reconstruction also remains distinct from prospective detection. A claimed warning at position t must be reproducible using only evidence available at or before t. Appending future events must not silently alter values represented as prospectively available.

Shared Behavioral Telemetry

The Current Evidence Run contains a shared behavioral telemetry substrate from which instruments and investigative surfaces receive their authorized measurements.

The telemetry layer may organize evidence-bound representations of:

• runtime continuity and displacement;

• recurrence and operational re-entry;

• semantic or objective drift;

• contradiction and pressure proxies;

• temporal density, coupling, lag, and shear;

• role and handoff dynamics;

• perturbation and response;

• containment and boundary support;

• regime development;

• candidate and confirmed transitions;

• observable outcome relationships;

• Recovery, relapse, or unresolved continuation;

• and source coverage and evidence integrity.

This shared layer replaces the need for a separate stability foundation that might appear to possess independent authority. Stability, continuity, basin, boundary, and recovery representations remain registered measurements or projections within the Current Evidence Run and, once issued, within the CER.

Identity coherence remains a continuity proxy, not proof of identity. Basin geometry belongs to the registered state representation, not to a hidden physical space inside the model. A risk score is not a calibrated probability unless external validation supports that interpretation.

Measurement and Signal Authority

A complex analytical system may produce hundreds of intermediate values, arrays, scores, labels, and display calculations. Without governance, any visible number can gradually be treated as a canonical measurement.

Aperture controls this through versioned Measurement and Signal Authority.

Every governed measurement or signal identifies:

• a stable name and family;

• the measurand or represented property;

• its authority and derivation class;

• eligible source fields;

• computation, transformation, or proxy relationship;

• temporal coordinate and window;

• evidence basis and dependencies;

• parameter and threshold configuration;

• treatment of missingness and conflict;

• eligible consumers;

• validation and implementation status;

• allowed claims;

• and prohibited interpretations.

The public architecture distinguishes three levels:

• Level 0 — Evidence-near telemetry: source-derived or minimally transformed runtime measurements.

• Level 1 — Derived runtime dynamics: registered relationships such as displacement, recurrence, coupling, continuity support, pressure proxies, or transition support.

• Level 2 — Instrument projections: higher-order worldline, regime, boundary, recovery, topology, and interaction views.

A Level 2 projection cannot be presented as a direct source observation. A Level 1 proxy cannot become an external fact because several interfaces repeat it. A Level 0 measurement still remains limited by source quality and method definition.

Registry counts are version-specific implementation facts, not part of the public identity of the architecture. When signals are consolidated, split, renamed, deprecated, or moved between levels, the applicable registry and method version must remain visible.

Contract-Bound Instruments

Aperture contains nine scientific instrument contracts:

• ∿ Seismo;

• τ Chronos;

• Δ Drift;

• Ξ Pressure;

• ⇄ Bridge;

• ⟳ Noesis;

• Ω Scope;

• Ψ Dynamics; and

• Φ Interferometer.

Each contract declares:

• which Current Evidence Run or issued CER objects it may read;

• which signal levels it may consume;

• which transformations it performs;

• which findings it may produce;

• how missing, conflicting, or unavailable evidence is handled;

• which temporal horizons apply;

• what it may display and export;

• and what it is prohibited from claiming.

Instrument findings are first-class projection objects. They retain the Current Evidence Run and Passport identity, authority level, evidence basis, signal dependencies, source spans, temporal coordinates, interpretation status, operational context, claim boundary, and creation lineage.

The visual design of an instrument may evolve without changing its scientific contract. Changing its eligible inputs, formula, threshold, classification rule, persistence requirement, or exported claim produces a new contract version.

Instrument agreement can demonstrate consistency across the implemented architecture. It does not constitute independent corroboration where instruments share source fields, signals, markers, or transformation dependencies.

Runtime Evidence Interpretation Matrix

The Runtime Evidence Interpretation Matrix, or REIM, is a versioned heuristic layer for organizing evidence-bound features into candidate interpretations and investigative routes.

Where enabled, it may distinguish conditions such as:

• stable operation;

• temporal formation;

• boundary formation;

• cumulative displacement;

• role-coordination pressure;

• tool-loop pressure;

• Recovery or re-entry;

• and multi-factor incidents.

REIM does not create evidence, modify telemetry, assign formal markers, or replace instrument findings. If no registered interpretation satisfies its rules, it may report only that Runtime Evidence was computed and that no qualifying interpretation is available.

That fallback is intentional. The architecture does not require every runtime to become an incident or every evidence object to produce a narrative.

REIM may be deterministic under a frozen version, but determinism does not make its weighted rules universally valid. It remains a disclosed interpretation layer whose inputs, thresholds, alternatives, and validation status must be inspectable.

Its recommendations route investigation. They do not authorize intervention, assign cause, or determine responsibility.

Runtime Reconstruction and Guided Investigation

Runtime Reconstruction brings the canonical spine, shared telemetry, temporal authority, instrument findings, optional REIM output, Passport identity, evidence lineage, and claim boundaries together in one investigation context.

The reconstruction may be organized into chapters addressing:

• source qualification and coverage;

• worldline and runtime formation;

• behavioral measurements;

• temporal organization;

• role and interaction topology;

• instrument findings;

• boundary and failure formation;

• replay and source inspection;

• claim lineage;

• and preservation readiness.

Guided Investigation provides the operator with relevant questions, inspection objectives, evidence status, source authority, confidence posture, alternative interpretations, and routes to supporting views.

The guide does not decide the case. It helps the operator move through unfamiliar evidence while preserving the differences among source observations, computational findings, human interpretation, unresolved questions, and formal claims.

Operator review may confirm, challenge, annotate, or withhold an interpretation under a declared process. Review does not convert a derived value into source evidence or strengthen the underlying claim beyond its evidence and method.

Runtime Evidence Passport Identity

The Runtime Evidence Passport identifies the CER issued from the Current Evidence Run.

It may preserve:

• source, runtime, and Passport identities;

• source and computation fingerprints;

• canonical event and frame coverage;

• computation, schema, contract, and build versions;

• role, model, toolchain, and operational disclosures when supplied;

• temporal and marker status;

• evidence and replay integrity;

• measurement and finding availability;

• claim boundaries;

• redaction and disclosure state;

• and preservation readiness.

The Passport contains deterministic and variable information.

Canonical source identity, configuration, frames, registered measurements, marker states, and method-bound outputs can belong to the governed evidence core. Generation time, browser-session activity, operator declarations, presentation state, and export context may vary.

Aperture therefore distinguishes the deterministic evidence identity from a particular Passport generation or export envelope.

The Passport identifies what was computed, whether the declared issuance requirements were satisfied, and which boundaries govern the certified record. It does not certify truth, source authenticity, intent, hidden state, consciousness, root cause, scientific validity, legal admissibility, or deployment safety.

Evidence Commons Preservation

Evidence Commons preservation carries the runtime beyond the live browser session.

A preservation bundle may contain:

• the machine-readable Certified Runtime Evidence Record and its Current Evidence Run lineage;

• qualified source identity and disclosure information;

• canonical runtime structures;

• measurement and signal lineage;

• instrument findings;

• human-readable investigation reports;

• temporal, role, validation, and method appendices;

• schemas and manifests;

• redaction and availability states;

• claim boundaries;

• and release information.

The objective is future challengeability.

A qualified reviewer should be able to:

• identify the source and its disclosed limits;

• inspect or recompute the canonical runtime where materials permit;

• locate the basis of a signal, marker, or finding;

• determine what an instrument was authorized to claim;

• distinguish computation from human interpretation;

• compare the result with another method or version;

• and see where evidence was missing, withheld, redacted, or not computable.

A PDF may present the investigation, but presentation alone is not a complete evidence artifact. Machine-readable objects, contracts, schemas, versions, and manifests preserve the authority path required for replay and challenge.

Browser-Local Execution and Scaling

Aperture performs its public computation locally in the browser without mandatory source upload to a remote analysis service.

This creates a practical foundation for independent research and sensitive operational review, but it also introduces memory, responsiveness, rendering, and export constraints.

The architecture addresses these constraints through:

• preflight qualification and compute estimation;

• declared operating bands;

• worker-based computation;

• essential-first evidence commit;

• progressive replay;

• deferred appendices;

• bounded visual hydration;

• sampled worldline rendering for large cases;

• and chunked export generation.

The governing principle remains:

Runtime Evidence is computed and preserved before a presentation strategy is selected.

Large runtimes may require reduced visual density, progressive inspection, or summary-first navigation. The canonical runtime and evidence-bearing measurements must not be silently truncated merely to satisfy the interface.

Evidence is preserved. Presentation adapts.

Browser-local operation does not eliminate every privacy or security risk. Imported content, browser extensions, local compromise, third-party dependencies, exports, retention behavior, and operator actions still require explicit controls and disclosure.

Conformance and Surface Integrity

Authority continuity must be tested across the entire system, not only inside the compute core.

Aperture can therefore be evaluated for conformance properties such as:

• all authoritative surfaces resolve to the same Current Evidence Run identity;

• formal markers remain positionally and semantically consistent;

• unavailable values are not rendered as zero;

• instrument inputs match their registered contracts;

• Operational Worlds do not alter canonical telemetry;

• retrospective outputs are not mislabeled as prospective;

• Passport fields resolve to the governed evidence core;

• exported findings retain source and method lineage;

• presentation sampling does not mutate the canonical runtime;

• and preservation artifacts retain the limitations shown during investigation.

Surface integrity matters because authority can be lost after correct computation. A valid marker can be mislabeled in a brief. A missing value can become a zero in a chart. An export can omit uncertainty. A responsive layout can sample points without disclosing that sampling occurred.

The architecture treats those failures as evidence-governance defects, not merely interface imperfections.

What the Reference Implementation Establishes

The reference implementation demonstrates that the proposed architecture has been translated into an operating computational system.

The system can:

• resolve and qualify heterogeneous sources;

• preserve roles and source provenance;

• construct canonical events and runtime frames;

• form a runtime spine and evidence-bearing worldline;

• compute registered longitudinal measurements;

• maintain shared temporal and marker authority;

• govern measurements and signals by versioned contracts;

• bind nine instruments to one Current Evidence Run;

• route investigation through bounded interpretation;

• identify the evidence record through a Passport;

• and preserve the reconstruction with its lineage, availability states, and claim limits.

This is stronger than a conceptual diagram. Each architectural proposition corresponds to executable objects, schemas, contracts, state transitions, conformance checks, and failure conditions.

It establishes that the architecture can operate under declared conditions. It does not establish universal validity for every measurement or scientific construct.

What the Reference Implementation Does Not Establish

The existence of working code does not prove that:

• Longitudinal Computational Dynamics is a complete account of computational intelligence;

• worldlines are universal natural invariants;

• symbolic time has universal units;

• Basin Exit predicts every failure;

• stability or continuity proxies correspond to hidden model entities;

• REIM generalizes across systems and domains;

• every threshold is calibrated;

• every regime corresponds to a natural kind;

• every source contains enough information for independent reconstruction;

• or every signal provides value beyond simpler baselines.

The implementation creates the conditions under which those propositions can be tested.

An external researcher can reject the terminology while still evaluating source integrity, canonicalization, role ablation, deterministic replay, temporal leakage, marker formation, signal dependence, instrument contracts, surface congruence, and preservation fidelity.

That independence is essential. The architecture should remain examinable even by investigators who do not accept its scientific interpretation in advance.

The Engineering Contribution

SubstrateX Aperture combines established lineages in logging, observability, event and process reconstruction, runtime verification, provenance, canonicalization, scientific instrumentation, visual analytics, forensic investigation, and AI governance.

Its distinctive engineering contribution is their integration around one source-bound runtime object and one non-amplifying authority path:

• role-bearing longitudinal reconstruction;

• one canonical runtime spine;

• multiple temporal coordinates and governed markers;

• shared behavioral telemetry;

• explicit measurement and signal authority;

• contract-bound instruments that cannot create independent evidence;

• evidence-bound interpretation and Guided Investigation;

• deterministic evidence identity separated from variable Passport metadata;

• bidirectional source-to-claim traceability;

• browser-local execution; and

• preservation of evidence, absences, uncertainty, interpretations, and limitations together.

SubstrateX Aperture™ implements a browser-local reference architecture for Evidence-Governed Runtime Reconstruction. Its defining achievement is not the number of metrics, instruments, or interfaces it contains, but the continuity of authority from source record to preserved claim.

The relationship to the wider public architecture is direct:

• SubstrateX® develops Runtime Evidence Infrastructure for long-horizon AI.

• Longitudinal Computational Behavior is the primary empirical object reconstructed by the system.

• Evidence-Governed Computation™ governs the authority path.

• Runtime reconstruction and behavioral telemetry are the core technical capabilities.

• Runtime Evidence is the source-bound evidentiary output.

• SubstrateX Aperture™ is the flagship Runtime Evidence Observatory and reference implementation.

• Evidence Commons preserves the resulting evidence for comparison and independent scrutiny.

Aperture makes the architecture executable. Authority continuity keeps every result accountable to its source, method, status, and limits.