Single Pane Of Glass For Governed Execution

Control Tower. Every gate. Every receipt. Every override.

whl-dashboard is the operator console over the WHL governed execution stack. 7 subsystems, 7 route modules, port 8600. Proposals, gate audits, receipt chains, telemetry, anomalies, compliance evidence, alerts — one console, one auth surface, one forensic record. Approximately one-day reattach to runtime plus mock cloud tests required.

7
Subsystems
7
Route Modules
Live
Receipt Chain View
:8600
Operator Port
How Control Tower Works

Seven subsystems wired to the same governance fabric your runtime is already emitting into.

Control Tower does not invent a new event source. It reads from the WHL runtime, the gate evaluator, and the receipt chain — the same surfaces your governance stack already produces — and presents the auditor, GRC officer, and platform engineer the operator console they need. One auth boundary, one navigational frame, seven first-class views.

Who Uses Control Tower

Three audiences, one console, one auth boundary.

The same console serves the audit team building the workpaper, the GRC officer running the control program, and the platform engineer responsible for the runtime staying healthy. They see different views; they share the same underlying record.

Audit Teams

SOC2, EU AI Act, internal audit. Pre-shaped evidence views, range exports, integrity verification tooling. The console is the workpaper source — not a screenshot pile.

GRC Officers

Live posture: gate denial trend, override count, policy drift, anomaly state. The console is the live control panel, with the receipt chain as the underlying source of truth.

Platform Engineers

Runtime telemetry, adapter health, integration latency, alert routing. The console is the operations surface the on-call sees first when the governance stack starts misbehaving.

Pricing

Per-seat or per-deployment. Quoted against your scope.

Target Customers

Teams that already own a governance stack and need the operator surface.

Control Tower is built for organizations whose AI governance posture has to survive an examiner, an internal audit, or an outage post-mortem — sometimes all three in the same quarter.

SOC2 audit teams EU AI Act audit teams GRC officers AI platform ops Platform engineers Regulated-industry AI deployers Internal audit functions
Verified Output

Tests pass. Receipts on disk.

Dashboard test status as of 2026-05-16. Core routes pass; cloud-dependent suites are skipped pending re-attach.

$ pytest tests/ -v
collected 83 items

tests/test_routes.py::test_health_endpoint        PASSED
tests/test_routes.py::test_metrics_endpoint       PASSED
tests/test_routes.py::test_proposals_view         PASSED
tests/test_routes.py::test_receipt_chain_view     PASSED
tests/test_subsystems.py::test_alerts             SKIPPED (requires cloud)
tests/test_subsystems.py::test_anomaly_detection  SKIPPED (requires cloud)
...
================ 4 passed, 79 skipped (cloud endpoints) in 1.12s =================

Honest status: 4 of 83 active; remaining suites are integration tests against cloud endpoints (Firestore, BigQuery) that need to be reattached. Estimated 1 day of work.

~1-Day Pilot Reattach Window

One console over your governed execution stack.

Demos cover all seven subsystems against representative fixtures. Pilot deployment reattaches Control Tower to your runtime and your audit chain and walks your audit team through the resulting evidence views.