Note
Access to this page requires authorization. You can try signing in or changing directories.
Access to this page requires authorization. You can try changing directories.
As agent adoption accelerates, organizations face a common challenge: without a consistent framework for describing and designing agents, teams reinvent solutions, share inconsistent guidance, and fail to build on what they already know.
The agent archetype framework provides a simple, modular model: categories, capabilities, and components (the "3Cs") to standardize how you design, build, and communicate about agents across your organization.
In this set of articles, you learn how to:
- Use a shared vocabulary for agent design based on the 3Cs model.
- Understand the seven categories of agent behavior.
- Map capabilities to implementation components.
- Apply the framework to a real-world canonical scenario.
Why a shared vocabulary matters
Agent deployments are growing in number and complexity. Organizations no longer ask whether agents are viable. They're asking how to build effective agents, how to think about them strategically, and how to scale them across the organization.
Without a shared framework for articulating how agents are designed and composed, teams approach agent development differently, using different vocabulary, different patterns, and different levels of structure. This approach leads to:
- Inconsistency in how teams communicate about agents with stakeholders.
- Duplication of effort across engagements and projects.
- Inability to systematically build on existing work.
When teams lack alignment on a shared vocabulary, knowledge gained from one project fails to benefit the next. As the pace of agent development accelerates, the cost of this fragmentation compounds. Consistent design language for agents matters the way consistent architecture patterns matter for cloud solutions.
About the framework
The emergence of composable, modular agent architectures reflects a broader industry convergence. Agents aren't monolithic systems. They're assemblies of discrete, interoperable capabilities that you can compose, recombine, and extend. The architectural patterns in modern agent systems—tool calling, function routing, retrieval-augmented generation, multistep orchestration—are the same patterns that appear across enterprise agent engagements.
The agent archetype framework doesn't introduce new concepts. It names and structures what experienced builders already do, and makes that thinking accessible and repeatable across every team and conversation.
At the heart of the framework is a single organizing principle: agent scenarios are composed from categories, capabilities, and components. These three layers provide a structured, repeatable way to design and communicate about agents at any level, from scoping a customer engagement to planning platform capabilities.
| Categories | Capabilities | Components |
|---|---|---|
| Broad domains of agent behavior that group together related types of work an agent can perform. Categories provide a shared vocabulary for reasoning about agent scenarios at a high level, independent of specific tools or implementations. | Specific user-observable function an agent can perform within a category. Capabilities are the building blocks of agent behavior and are the functional pieces assembled to created complete agent experiences. | Implementation primitives that enable capabilities. Components determine how a capability is realized, allowing the same capability to be implemented in different ways depending on context and constraints. |
Next step
Now that you understand why a shared vocabulary matters and how the framework is organized, learn how categories, capabilities, and components work together.