Overview
The controls that make agentic systems operable
The governance plane is composed of a small set of controls that determine how agentic systems behave in production.
Each control exists to address a specific operational constraint. Together, they define whether an agent-based workflow can be run safely, predictably, and over time.
What follows is an overview. Each capability is described in detail on its own page.
POLICIES
Governing what agents are allowed to do
Policies define the rules under which agentic work may proceed. They determine what actions are permitted, when approvals are required, and how exceptions are handled.
Because policies are applied centrally, governance remains consistent even as workflows change.
TELEMETRY
Observing behavior as systems run
Telemetry provides continuous visibility into how agent systems are performing in production. It surfaces patterns, drift, and anomalies without interfering with execution.
Operational insight is available as systems run, not only after failures occur.
TESTING
Evaluating changes before they reach production
Testing allows teams to evaluate agent behavior and governance changes in controlled environments before deployment.
This reduces the risk of introducing unvalidated behavior into live systems.
registry
The Registry defines the agents, workflows, and tools that are allowed to exist and execute in the system. It makes execution structure explicit rather than inferred.
Teams can see how systems are actually composed and verify that execution matches intended design.
AGENT COMPOSer
Composing agents from intent and existing logic
Agent Composer turns high-level intent or existing agent logic into governed, deployable agents inside Waxell.
This allows teams to introduce new and migrated agents safely without rewriting systems or bypassing governance.
SIGNAL & DOMAIN
Controlling how systems exchange data
Signal and Domain define the inbound and outbound interfaces between production systems and the Waxell runtime. They make data flow and action execution explicit rather than implicit.
Teams can introduce autonomous workflows without exposing internal systems or losing traceability.
COMMAND LINE INTERFACE
Operating agent systems from the command line
The Waxell CLI provides a unified interface for observing, testing, and operating agent systems directly against the runtime.
This allows developers to act on live systems without switching tools or bypassing governance.
How these work together
Each capability addresses a different aspect of operability. None is sufficient on its own.
Budgets and policies define boundaries.
Executions and telemetry provide visibility.
Testing reduces risk as systems evolve.
Together, they form a governance plane that allows agentic systems to be treated as production infrastructure rather than experimental tooling.
