What are agent workflow patterns? (And when a workflow is enough)

Agent workflow patterns are the five shapes teams reuse when wiring model calls, tools and control flow together: prompt chaining, routing, parallelization, orchestrator-workers and evaluator-optimizer. Three of them are branch logic you can already draw in Zapier or n8n. Only two need a runtime that picks the next step while the job runs.

The five patterns

Anthropic's "Building Effective Agents" names five, and most frameworks since have renamed the same list.

PatternWhat runsPath decided
Prompt chainingLLM calls in series, each consuming the last outputAt build time
RoutingA classifier sends the input down one branchAt build time
ParallelizationThe same step run several times and voted, or split into sectionsAt build time
Orchestrator-workersA planner invents subtasks while the job runsAt run time
Evaluator-optimizerA second model critiques the first and loopsAt run time

Three of the five are branch logic you already draw

Chaining, routing and sectioning are control flow with a model in the middle. If the path is knowable while you build it, draw it in Zapier or n8n, because a drawn path fails somewhere you can point at. n8n's AI Agent node concedes the same split: since version 1.82.0 it runs only as a Tools Agent and refuses to start without at least one tool connected. Its job is picking a tool at run time, not walking a route you already mapped.

When the run-time patterns earn it

The recurring complaint in Hacker News threads on agent projects is that past a certain number of steps the agent loses context, repeats steps, or goes off the rails. A chain terminates by construction. An open loop needs an iteration cap and a human exit. Rebuild a flow as orchestrator-workers or evaluator-optimizer only when the steps cannot be listed before the run starts. Everything else stays a workflow.

Last updated: May 20, 2026

Build your agent team in 30 seconds.

Build agent teams that work along with your team. Free to start, no card required.