Controlled public proof — not a claim of customer production endpoint enforcement.

Interactive public proof

See what happens before an AI proposal becomes organizational consequence.

Choose a public-safe case. Inspect the authoritative ALLOW, BLOCK, or ESCALATE result, release control, consequence proof, route maturity, controlled-batch posture, and browser-side packet verification.

Controlled event-level proof is shown. Production L5 route completeness and non-bypassability require customer-specific integration evidence.

Current controlled batch + illustrative scenariosBehavioral public projectionBrowser-side SHA-256 checksNot customer production enforcement
1 · Decision Explorer

Choose the movement the AI attempted to cause.

Inspect the Decision API result for a public-safe case. Scores do not authorize release.

Loading

Loading proof case…

Attempted action
Potential protected effect
Decision API result
LOADING
Release
Release authorized
Human review
Protected effect formed
Commit
Proof scope

Public Explorer data is a behavioral projection only. Detailed packets are available under technical diligence.

2 · Lifecycle

Proposed action to consequence evidence.

    3 · Controlled batch

    Current controlled batch + additional illustrative scenarios.

    Five governed movements from the current public batch, plus three illustrative cybersecurity scenarios that are not members of that batch.

    Aggregate decision
    Aggregate release
    Human attention
    Proof postureControlled public proof

    Showing all controlled cases.

    CaseAttempted actionDecisionReleaseEffect formed

    Illustrative cybersecurity scenarios — not AgenTrust TRACE cases and not members of the current controlled batch.

    DemonstrationAttempted actionDecisionReleaseEffect formed
    4 · Browser-side verification

    Do not take the packet on faith.

    This browser checks SHA-256 hashes of the public Explorer files and compares the public receipt to the selected case. It is not an external digital signature.

    Verification resultNot run0 files checked · 0 verified
    • Checksum manifestWaiting
    • Public file hashesWaiting
    • Receipt / case parityWaiting
    Public capability boundary

    What this Explorer proves—and what it does not claim.

    Shown here

    • ALLOW / BLOCK / ESCALATE
    • Release authorized or withheld
    • Whether a protected effect formed
    • Human review required or not
    • Controlled public proof scope

    Requires customer integration

    • Production endpoint enforcement
    • Fail-closed route evidence
    • Reviewer and outcome closure

    Not claimed

    • Universal production L5
    • Hardware attestation
    • AgenTrust Verified
    Next step

    Which AI-proposed movement should OntoGuard govern first?