Human-on-the-Loop

A human supervises agents and intervenes only when thresholds are crossed.

Why it matters

Most agent actions don't need approval. Human-on-the-loop lets agents work freely within safe boundaries.

In practice

Our 'supervised' agents send daily summary reports via Telegram. The human reviews and intervenes only if needed.

Related terms

Human-in-the-Loop (HITL)

A human reviews and approves specific agent actions before they execute.

Governance

Policies, controls, and processes that ensure responsible agent behavior in production.

Heartbeat

A scheduled check-in where an agent wakes, reviews its tasks, acts, and reports its status.

Back to glossary