◈ AWS AGENTCORE BLUEPRINTS
PART IV · GOVERNANCE AND TRANSACTIONS · CHAPTER 23 / 36
TOCCONSULTING.FR
AMAZON BEDROCK / SERVICE DETAIL S10

AGENTCORE POLICY

CHAPTER 23 / 36
IN ONE LINE a hard checkpoint in front of every tool call
REV 2026.07

An agent decides for itself which tools to call, which is powerful and a little scary: it might misread a rule, or be tricked into doing something it should not. Policy puts an automatic checkpoint in front of every tool call, outside the agent, that allows or blocks it based on rules you set, so the agent literally cannot cross the lines you draw.

New to these words? Read chapter 02 "Key terms" (tool, MCP). See also chapter 11 "Gateway".
[ 1 ] WHAT IT IS

Policy is a checkpoint that sits at the Gateway and inspects every tool call before it runs. You write rules (in plain English or in Cedar, AWS's open-source rules language), store them in a policy engine, and attach the engine to a gateway. From then on, every request is checked against the rules and allowed or denied, the same way every time.

THE PROBLEM IT SOLVES
Rules buried in agent code can be misread, or bypassed by manipulating the agent.
THE FIX
The check lives outside the agent, so it is deterministic and cannot be argued with.
[ 2 ] HOW A TOOL CALL IS CHECKED
STEP 01

AGENT ACTS

The agent tries to call a tool, say "process a $2000 refund".

==>
STEP 02

GATEWAY INTERCEPTS

The request hits the gateway, where the policy engine sits.

==>
STEP 03

EVALUATE

The engine checks the tool, who is asking, and the inputs against your rules.

==>
STEP 04

ALLOW OR DENY

Permitted calls go through; the rest are blocked. Default is deny; any "forbid" wins.

==>
STEP 05

LOG

The decision is recorded in CloudWatch for audit.

[ 3 ] TWO WAYS TO WRITE A RULE
PLAIN ENGLISH easiest
Describe the rule in words: "only allow refunds under $1000".
AgentCore turns it into Cedar and checks it against the tool.
It even flags rules that are too loose, too strict, or impossible.
CEDAR precise
Write the rule directly for exact control.
permit or forbid an action on a tool...
...with a condition on the inputs (for example, amount < 1000).
[ 4 ] WHY IT IS SAFE, AND HOW IT DIFFERS

Because the check is at the boundary, outside the agent, it is deterministic: the same input always gives the same decision, and a clever prompt cannot talk it out of a rule. In ENFORCE mode, everything is denied unless a rule permits it, any forbid wins, and every decision is logged.

NOT THE SAME AS Policy checks tool calls (what the agent may do). Bedrock Guardrails checks model text (what the model may say). IAM checks AWS permissions (which AWS resources you may touch). Use them together.
Plain definitions: chapter 02 "Key terms". Hands-on setup: chapter 24 "Policy, practical".
SRC: AWS AgentCore Developer Guide (policy, getting started, core concepts). DRAWN 2026.06, REV 2026.07.26
◄ PREVIOUS · CH 22
AgentCore Evaluations, Practical
NEXT · CH 24 ►
AgentCore Policy, Practical
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