What you can build

Skills, Integration Flows, Managed Agents, and MCP context bundles, and when each one fits.

Once you know what to solve, choose the shape it should take. Four primitives cover most of what teams build on Klarity.

Skill

The simplest primitive: an instruction manual, or recipe, for an agent. Governable, composable, and small. A skill runs in any AI tool and is interoperable across providers. A skill is the canonical simple agent.

Integration Flow

Deterministic, rule-based automation on a fixed, predictable, auditable path. It does not use AI reasoning, so it is not an agent. Best where the process is stable and every step must be explainable.

Managed Agent

An AI agent that reasons and plans its own path under a human gate. It handles variable inputs a fixed flow can’t. Chained across multiple tasks and tools, it becomes a compound agent.

MCP context bundle

Packaged Company-Brain context that grounds an external AI tool, custom GPT, or code-generation build.

Simple and compound

Simple versus compound agents is the line between the two tracks. A simple agent does one thing, which is a skill. A compound agent chains multiple tasks and tools, and needs an SDK.

	Simple	Compound
What it does	One task	Chains tasks and tools
Built on	The non-technical track	The technical track
Typical primitive	Skill	Managed Agent

Compound agents scale further as orchestrated or multi-agent systems, custom SDK-built agents, and, as a horizon, fully autonomous agents.

Oversight

Set oversight by stakes. Every build carries a human gate, and you graduate from more review to more autonomy as evaluation results prove out.

Next, see Where you build for the surfaces each of these is built on.