Glossary

Control Tower

The live operations console for the whole agent fleet — watch, gate, pause, or kill.

category ▸ Operations

audit/verify
#101hash ✓#102hash ✓#103hash ✓#104hash ✓#105hash ✓
head 0x9f3a…c1ok: true
verifyChain ▸ chained SHA-256 · signed receipts
What it means

Control Tower, in plain language.

A control tower is the single pane of glass over every agent in production: live runs, costs, denials, reliability, identities needing review, and tool-server status, aggregated so an operator can see the whole fleet at once. It is the supervisory layer that turns a sprawl of autonomous agents into something a human can actually oversee.

More than a dashboard, a control tower is a place to act. Seeing a degrading or misbehaving agent is only useful if you can intervene immediately — pause it, tighten its oversight, or stop it entirely — from the same console.

In Cortex

How Cortex implements it.

This term isn't abstract here — it maps to a real capability in the runtime. Here is exactly how Cortex enforces or relates to it.

Operations

Cortex's Control Tower aggregates fleet-wide signal — average reliability and below-threshold counts, policy rules and tests, budget status, MCP server health, and identity recertification — into one overview with live KPIs and CSV export.

It is also where you intervene: pause a degrading agent, gate its activity, or kill it, with every supervisory action sealed into the Trust Ledger. This delivers the live supervision that frameworks like FINRA Rule 3110 expect.

See Control Tower enforced, not just defined.

Book a walkthrough and watch the controls in this glossary return real verdicts, seal real evidence, and trace every fact back to its source.