The gate between machine intent and real-world execution.
ASE evaluates whether this AI-assisted action has sufficient authority, policy alignment, and evidence to proceed.
ASE Gate is the runtime mode of the Auditome Sovereign Engine — designed to evaluate whether a machine or system action may proceed, before consequences occur, and to return a decision with a traceable, signed record. Live in demo. Production deployment coming — early access open.
What ASE Gate Does
Systems submit an intent to ASE Gate. It is designed to evaluate that intent against policy, authority, risk, and execution rules and return a recorded decision with stated reasons before anything runs.
Receipt issued. Action authorized for this request only.
Required authority, policy, or evidence is missing. ASE failed closed.
Policy requires human review. No execution receipt is issued. A signed escalation receipt is produced — the precondition for the governed human review path.
Every decision includes
Why ASE Gate Exists
Most systems
Can tell you what happened. Logs are recorded after execution. Audits run after the fact. Accountability arrives after consequences.
ASE Gate
Helps determine whether an action should have happened before consequences occur. Authorization, accountability, and approval boundaries are designed to be enforced at the execution boundary, not reconstructed afterward.
Should this principal be allowed to perform this action?
Every decision has an owner, a trace, and a record.
Define which actions require human sign-off before execution.
Enforce rules at the point of action, not the point of review.
How It Works
The system submits a structured intent. ASE Gate is designed to evaluate it and return a decision synchronously. Execution only proceeds if the decision is approved.
{
"decision": "approved",
"trace_id": "ase_tr_...",
"risk_score": 0.3,
"reasons": [],
"audit_ref": "aud_..."
}Every decision includes a trace_id and audit_ref for downstream traceability. This is the live response shape from the KRONYQL demo — production deployment coming.
One engine, two modes
ASE Gate is not a separate product. Diagnostic and Gate are two modes of the Auditome Sovereign Engine — the same engine, applied at two different points.
ASE Foundation Diagnostic
ASE inspects an AI-built or AI-assisted codebase read-only and reports what it can and cannot prove.
ASE Gate
ASE sits in the execution path, evaluates intent against policy and authority, and returns approved, rejected, or requires_human_review with a signed decision record.
The diagnostic shows where your system cannot prove itself. The gate is designed to enforce proof at the point of action.
Gate Verdict Reference
Action may proceed.
Receipt issued. Action authorized for this request only. A signed, durable receipt is returned and logged. Downstream execution is permitted.
Action is blocked.
Required authority, policy, or evidence is missing. ASE failed closed. No receipt is issued. Reasons are logged. Execution does not proceed.
Escalated. Pending human approval.
Policy requires human review. No execution receipt was issued. The decision is held pending an authorized human approver. Audit record is created.
The architecture is defined. The three decision states — approved, rejected, requires_human_review — are the designed interface. The policy evaluation model, trace record format, and audit persistence layer are specified.
ASE Gate does not have a live endpoint yet. If you need to evaluate AI actions against policy before they execute, register your interest and we will contact you when access is available.
Register Interest