Official Source· AWS Artificial Intelligence· AI· Published Nov 6, 2026

AWS maps agentic primitives in Amazon Quick, from triggers to workflows

In a November 6, 2026 post, AWS groups Amazon Quick’s agentic primitives into triggers, agents, capabilities, and learned context, using four questions—When, What, Who, and How—to explain their roles.

AWS maps agentic primitives in Amazon Quick, from triggers to workflows

Full Report

AWS published a taxonomy of agentic primitives in Amazon Quick on November 6, 2026, organizing them into triggers, agents, capabilities, and learned context. The framework breaks agentic work into four questions: when it runs, what it does, who performs it, and how it is carried out. A trigger, in this model, both activates work and carries task instructions, covering the “When” and “What.”

For a request to monitor Slack and send a morning digest, the post says the usual approach is to add a trigger to an existing agent rather than create another agent or workflow. Capabilities describe how an agent acts, ranging from fully agentic to fully deterministic execution—a distinction that reflects how much latitude the agent has.

AWS distinguishes tools from workflows as well. Tools are atomic operations with typed schemas, such as sending email, querying a database, or reading a file. Workflows coordinate actions in a fixed order through declared steps, control flow, branching, and error handling, with decision logic in code. The post says skills can capture, version, and share institutional expertise, while workflows offer auditability, step-wise visibility, and deterministic coordination across multiple services.

The taxonomy follows AWS’s June 2026 announcement of autonomous agents in Amazon Quick, adding a framework for breaking down agent tasks and choosing how to implement them. For teams designing automations, the distinctions clarify when to extend an existing agent, when to use a structured workflow, and how much autonomy a task may require.

Why This Matters

The framework separates triggers, agent behavior, tools, and fixed workflows, giving Amazon Quick users a more concrete way to decide whether a task belongs with an existing agent or in a step-wise, auditable process.

Evidence

Start with the primary evidence, then review supporting sources and context.

1 items

Primary Evidence

1
  • OfficialAWS Artificial IntelligenceDocumentOriginal

    Taxonomy of agentic primitives in Amazon Quick

    Thomas Delteil

    “As we built Amazon Quick, its agents grew from answering questions to taking actions, and the same primitives kept getting rebuilt under different names. This post shares the taxonomy we converged on — triggers, agents, capabilities, and learned context — found by decomposing agentic work into four plain questions: When, What, Who, and How.”

Recommended signals

Recommended by brand, topic, and recent related coverage

No more recommended signals right now

Official statement

Open original source