AWS Builder Center

A Beginner's AWS Security Monitoring Workflow: CloudTrail, GuardDuty, and Security Hub

A practical introduction to three AWS security services and how they can work together to help builders investigate activity, detect potential threats, and understand security findings.

A Beginner's AWS Security Monitoring Workflow

When I started learning AWS security, one question seemed more important than memorizing individual services:
If something suspicious happens in an AWS account, where should I look first?
A useful way to approach this is to understand three different layers of visibility:
CloudTrail → GuardDuty → Security Hub
They solve different problems, but together they provide a simple mental model for investigating security activity in AWS.

1. CloudTrail: Understand What Happened

AWS CloudTrail records AWS API activity.
For an investigation, CloudTrail can help answer questions such as:
  • Who performed an action?
  • What API operation was called?
  • When did it happen?
  • What source IP address was involved?
  • What resource or parameters were used?
For example, imagine an IAM policy is unexpectedly modified.
Instead of only asking, "Why is this policy different?", we can start with CloudTrail and investigate the API activity that caused the change.
A simple investigation mindset is:
Identity → Action → Source → Time → Resource
This turns raw cloud activity into a timeline that is much easier to understand.

2. GuardDuty: Identify Suspicious Behavior

Reviewing every event manually does not scale.
Amazon GuardDuty continuously analyzes supported AWS data sources and uses threat intelligence and machine learning to identify potentially malicious or unauthorized activity.
Examples can include:
  • Suspicious use of AWS credentials
  • Communication with known malicious infrastructure
  • Unusual API activity
  • Potential privilege escalation
  • Malware or cryptocurrency-mining related behavior
This means CloudTrail and GuardDuty answer slightly different questions.
CloudTrail: What happened?
GuardDuty: Does this activity look suspicious?
That distinction is useful when learning cloud incident investigation.

3. Security Hub: Bring Findings Together

As an AWS environment grows, security findings may come from multiple services.
AWS Security Hub and Security Hub CSPM provide centralized security visibility and can help builders review and prioritize security issues instead of checking each service independently.
A useful workflow is:
  1. Review the security finding.
  1. Identify the affected account and resource.
  1. Check the severity and finding details.
  1. Pivot to the underlying activity and logs.
  1. Determine whether the behavior is expected or suspicious.
  1. Take remediation action when necessary.
Security findings should be treated as the beginning of an investigation, not automatically as proof that an attack succeeded.

4. Putting the Workflow Together

Consider this simplified scenario:
A GuardDuty finding reports unusual API activity associated with AWS credentials.
I would investigate it using this workflow:

Step 1 — Read the finding

Identify:
  • Finding type
  • Severity
  • AWS account
  • Region
  • Resource
  • Principal or identity involved

Step 2 — Investigate CloudTrail

Search around the same timestamp and identity.
Look for:
  • API calls before and after the alert
  • Source IP addresses
  • Authentication activity
  • IAM changes
  • Resource creation or modification

Step 3 — Establish Context

Ask:
  • Is this normal activity for this identity?
  • Has the source IP been used before?
  • Is the action expected for the user's role?
  • Were additional resources accessed?
  • Did permissions change?

Step 4 — Decide the Outcome

The investigation may result in something like:
Benign / Expected Activity
The behavior matches an authorized administrative or automation activity.
Needs More Investigation
There is not enough evidence to determine whether the activity is legitimate.
Suspicious / Confirmed Security Event
The activity is unauthorized or additional evidence supports malicious behavior.

5. A Simple Mental Model

For beginners, I think the easiest way to remember these services is:
CloudTrail = Activity
GuardDuty = Detection
Security Hub = Findings and Prioritization
It is not a complete cloud security architecture, but it provides a good starting point for learning how security monitoring works in AWS.

Final Thoughts

Learning individual AWS services is useful, but learning how information flows between services is even more valuable.
Instead of asking only:
"What does this AWS service do?"
Try asking:
"How would I use this service during a real investigation?"
That small change in perspective makes cloud security much easier to connect with practical scenarios.
If you work with AWS security or are currently learning it, what service do you usually check first when investigating suspicious cloud activity?
Any opinions in this article are those of the individual author and may not reflect the opinions of AWS.
Enjoyed reading this content? Let the author know!

Your likes, comments, shares, and saves help creators reach more builders.

Loading recommendations

Loading article