◈ AWS AGENTCORE BLUEPRINTS
PART V · MULTI-AGENT AND ARCHITECTURE · CHAPTER 31 / 36
TOCCONSULTING.FR
AMAZON BEDROCK / PART V, MULTI-AGENT AND ARCHITECTURE

ARCHITECTURE PATTERNS

CHAPTER 31 / 36
IN ONE LINE how to structure your agents
REV 2026.07

Before you build, two questions decide your architecture: should this be one agent or several, and what triggers it, a person chatting, or an event? This is a decision guide: when to split into specialists, the three ways to organize them, and how event-driven agents run without a human.

New to these words? Read chapter 02 "Key terms". Agents working together: chapter 29. Serving many customers: chapter 32.
[ 1 ] FIRST QUESTION: ONE AGENT, OR MANY?

Start with one narrowly scoped agent. Split into several only when one agent gets unwieldy: the prompt grows complex, tool selection gets unreliable, performance drops. The test is like a team, you do not hire one person for sales, engineering, support, and finance. Instead of one agent doing thirty tasks, build three that each do ten related ones: clearer instructions, simpler tools, easier to debug.

KEEP IT ONE WHEN
The job is focused and the tool set is small.
SPLIT INTO MANY WHEN
Responsibilities pile up and the single prompt becomes a mess.
[ 2 ] THREE WAYS TO ORGANIZE MANY AGENTS
SEQUENTIAL
WHEN STEPS ARE FIXED

Tasks have a natural order. Agent 1 retrieves, agent 2 analyzes, agent 3 writes the report. A pipeline.

HIERARCHICAL (SUPERVISOR)
WHEN YOU NEED ROUTING

A supervisor reads the request, works out the intent, and delegates to the right specialist. A widely used pattern.

PEER-TO-PEER
WHEN WORK IS OPEN-ENDED

Agents collaborate dynamically, with no central boss, handing off to whoever fits the moment.

[ 3 ] WHAT STARTS THE AGENT: A PERSON, OR AN EVENT?
STEP 01

EVENT

A file lands in S3, or an email arrives. No one is chatting.

==>
STEP 02

ROUTE

Amazon EventBridge catches the event.

==>
STEP 03

TRIGGER

A Lambda invokes the agent on Runtime.

==>
STEP 04

WORK

The agent processes it and scores its confidence.

==>
STEP 05

ACT

It routes the result: act automatically, or send for human review.

This is the event-driven pattern: the agent runs in the background, triggered by what happens, not by a conversation. (Example from the samples: an insurance-claims pipeline.)
[ 4 ] DO NOT CONFUSE PROTOCOLS WITH PATTERNS
PROTOCOLS how they talk
A2A, MCP, HTTP. The wire format, the API contract. The infrastructure layer.
PATTERNS how they organize
Sequential, hierarchical, peer-to-peer. The workflow and coordination. The architecture layer.
You can use the same protocol with different patterns, so keep them separate and do not tie your business logic to your plumbing. And when one agent hands work to another, the next one needs the context, AgentCore Memory carries it. Watch the handoffs: that is where most failures happen.
Plain definitions: chapter 02. A2A and multi-agent code: chapter 29. Multi-tenant: chapter 32.
SRC: AWS ML Blog (AI agents in enterprises, best practices with AgentCore) + awslabs samples (event-driven-claims-agent). Guidance. DRAWN 2026.06, REV 2026.07.26
◄ PREVIOUS · CH 30
A2A and Multi-Agent, Practical
NEXT · CH 32 ►
Multi-Tenant Agents, Plain Language
AWS AGENTCORE BLUEPRINTS · CONTENTS
Back to tocconsulting.fr Cover & Table of Contents
PART I · FOUNDATIONS
01AgentCore Overview, Plain LanguageCONCEPT02AgentCore, Key Terms in Plain LanguageCONCEPT03AgentCore, Common Mistakes ExplainedCONCEPT04Inside an AgentCore Agent, BlueprintCONCEPT05Harness, Practical 1, Strands PathPRACTICAL06Harness, Practical 2, Managed HarnessPRACTICAL
PART II · CORE SERVICES
07AgentCore Runtime, Plain LanguageCONCEPT08AgentCore Runtime, PracticalPRACTICAL09AgentCore Memory, Plain LanguageCONCEPT10AgentCore Memory, PracticalPRACTICAL11AgentCore Gateway, Plain LanguageCONCEPT12AgentCore Gateway, PracticalPRACTICAL13AgentCore Identity, Plain LanguageCONCEPT14AgentCore Identity, PracticalPRACTICAL
PART III · TOOLS AND OPERATIONS
15AgentCore Code Interpreter, Plain LanguageCONCEPT16AgentCore Code Interpreter, PracticalPRACTICAL17AgentCore Browser, Plain LanguageCONCEPT18AgentCore Browser, PracticalPRACTICAL19AgentCore Observability, Plain LanguageCONCEPT20AgentCore Observability, PracticalPRACTICAL21AgentCore Evaluations, Plain LanguageCONCEPT22AgentCore Evaluations, PracticalPRACTICAL
PART IV · GOVERNANCE AND TRANSACTIONS
23AgentCore Policy, Plain LanguageCONCEPT24AgentCore Policy, PracticalPRACTICAL25AgentCore Registry, Plain LanguageCONCEPT26AgentCore Registry, PracticalPRACTICAL27AgentCore Payments, Plain LanguageCONCEPT28AgentCore Payments, PracticalPRACTICAL
PART V · MULTI-AGENT AND ARCHITECTURE
29A2A and Multi-Agent, Plain LanguageCONCEPT30A2A and Multi-Agent, PracticalPRACTICAL31Architecture Patterns, Plain LanguageCONCEPT32Multi-Tenant Agents, Plain LanguageCONCEPT33Agent Security and Threat Model, Plain LanguageCONCEPT34Governance at Org Scale, Plain LanguageCONCEPT35Governance at Org Scale, PracticalPRACTICAL36Multi-Region and Distribution, Plain LanguageCONCEPT