Fleet 2.6.0 is out.See what's new →
FleetFleet
Glossary

AI Agent Control Plane

An AI agent control plane is the centralized management layer for an organization's autonomous agents — the single surface where agents are registered, configured, started and stopped, governed, monitored, and audited, separate from the data plane where the agents actually do their work.

The control-plane / data-plane split is borrowed from networking and infrastructure. The data plane is where the work happens — in an agent system, that is each agent executing tasks: reading files, running tests, opening pull requests. The control plane is the layer that decides which agents exist, what they are allowed to do, when they run, and how their behavior is recorded. Keeping the two separate is what lets an organization scale from a handful of agents to a managed fleet without losing oversight.

A control plane answers operational questions that no single agent can answer about itself: Which agents are registered, and who owns each one? What is each agent's role and scope? Is an agent currently running, idle, or quarantined? What has each agent done, and under which governing rules? Without a control plane these answers are scattered across laptops, shell scripts, and individual API dashboards — the condition that produces shadow AI and agent sprawl.

In practice a control plane provides agent registration and identity, lifecycle commands (start, stop, schedule), policy and governance (roles, approval gates, run-time budgets, risk controls), and observability (a shared audit trail and status view). It is the difference between running agents and managing a fleet of them.

How this relates to Fleet

Fleet is a self-hosted control plane for fleets of Claude Code agents. Agents are declared in checked-in config (org.yaml or .fleet/config.yaml), started and stopped through one CLI, governed by role definitions, approval gates, run-time budgets, and a risk model, and recorded in a shared audit trail surfaced by fleet log. The agents themselves — the data plane — run as Claude Code sessions; Fleet is the control plane that registers, governs, and audits them. A hosted dashboard layer at app.fleetctl.ai adds cross-fleet visibility and governance policy on top of the self-hosted binary.

Frequently asked questions

What is the difference between an agent control plane and agent orchestration?

Orchestration is one function of a control plane: sequencing and routing work between agents at runtime. A control plane is broader — it also covers registration and identity, lifecycle management (start/stop/schedule), governance and policy, and the audit trail. You can orchestrate agents without a real control plane, but you cannot govern a fleet at scale without one.

Does an AI agent control plane have to be self-hosted?

No. Control planes exist as hosted SaaS, as self-hosted software, and as hybrids. The trade-off is data residency and control versus operational convenience. Fleet's control plane is self-hosted as a single binary so your source code stays on your own infrastructure, with an optional hosted dashboard for cross-fleet visibility and governance.

Run your first agent fleet

One binary. Five minutes. See every agent, coordinate every handoff, and keep a full audit trail of what your fleet did.