ESCALATED · op_5f0e2c79 — REGENT decision record

Decision record

ESCALATED

Operation: op_5f0e2c79
Request: req_5f0e2c79
Tenant: UniversalHealthCare.ai
Subject: UHC-PT-00838synthetic

tamper-evident with governed redaction

Outreach blocked. The named human escalation target was alerted.

A crisis flag stopped the operation before the judgment engine ran. Neither the primary target nor the clinical duty desk acknowledged inside their SLAs, so the named operator alarm was raised and retries until a human acknowledges it.

Timing

Received by Regent

03:04:05 UTC

16 Aug 2026 · 03:04:05 UTC · computed by Regent

Resolved

03:04:12 UTC

Operation reached its terminal state

Processing latency

7.0 s

Receive → resolve, on Regent's clock

Tenant snapshot timestamp

03:03:58 UTC

tenant-asserted · not the basis of the freshness guarantee

Regent's freshness guarantee begins when Regent receives the push. The snapshot timestamp inside the request is tenant-asserted metadata and is never the basis of that guarantee; data age before the push is the tenant's pipeline responsibility.

Versions in effect

Policy versionuhc_reengagement.v3

Model versionnone — the judgment engine was not invoked

Channelwhatsapp

IntentOvernight welfare check

AI judgment

AI judgment — NOT RUN · blocked by pre-agent hard stop

When a crisis flag is present in the accepted request context, the pre-agent hard stop resolves the escalation before the judgment engine is invoked. No outreach is proposed, none is sent, and the named escalation path executes.

Context · pre-agent hard stops

Contactability
Active WhatsApp channel.
evidence: computed by Regent
PASS
Crisis flag / global opt-out
Crisis flag present in the accepted snapshot. Non-overridable hard stop.
evidence: tenant-asserted
ESCALATE

Judgment · AI proposal

AI judgment — NOT RUN
Blocked by the pre-agent hard stop. No proposal exists.
evidence: computed by Regent
NOT RUN

Governance · rules run after the proposal

Post-proposal governance
Not reached — there is no proposal to evaluate.
evidence: computed by Regent
NOT RUN

Action · dispatch

Named escalation path
Primary target and fallback target both failed to acknowledge. Operator alarm raised at 03:34:14 UTC: a Sentry alert on the wired project plus a WhatsApp message to the named operator, backed by a durable alarm row that retries until acknowledged.
evidence: computed by Regent
ESCALATE

Dispatch state

Providernone

Provider acceptance· No dispatch attempted

Escalation

Named targetDr. Nour Khalil · on-call clinical lead

Stateoperator alarm

Acknowledgment SLA15 minutes

Acknowledgednot yet

Non-overridable at decision time. There is no operator bypass.

Evidence

Trace idtrace_5f0e2c79

Integrity statusdemo

Signed checkpointdemo-checkpoint-0182

Data sourcedemo · synthetic fixture

Exact inputs, policy versions, model version, gates and recorded outcome are preserved and integrity-verifiable on demand. Inspect reconstructs the recorded decision context — the inputs, versions, gates and outcome exactly as they were. It does not re-run the model or promise an identical regenerated answer.