Lion Cipher Lab
Set up AWS GuardDuty threat detection (your own free-tier account, sample findings)
- Cloud Security
- Beginner
- ~45 min
- Free · no account required
Why it matters
In the cloud, the danger is not only a bad setting sitting still -- it is malicious ACTIVITY happening right now: stolen credentials being used from a strange country, an instance quietly talking to a known-bad server, someone probing your account. You cannot watch for that by hand. Threat detection is the discipline of having a service watch the account's activity for you and raise an ALERT the moment something looks like an attack. On AWS the built-in answer is Amazon GuardDuty: you switch it on, and it continuously reads the account's own activity logs -- the record of every API call…
What you'll learn
- what cloud threat detection is: a service that watches your account's activity and raises an alert when something looks malicious
- what Amazon GuardDuty is and where its signal comes from -- the foundational data sources CloudTrail, VPC flow logs, and DNS query logs -- with no logging for you to set up
- enabling GuardDuty in your own free-tier account, one Region at a time, and understanding the 30-day free trial
- generating SAMPLE findings with GuardDuty's built-in button so you can learn the console safely, with no real attack and no risk
- reading a finding: the [SAMPLE] prefix, the finding type, the affected resource, and the full JSON detail
- what severity means -- GuardDuty's 1.0-10.0 scale split into Critical / High / Medium / Low -- and how it drives what you do first (triage)
- routing findings to a human: sending GuardDuty alerts to your email through EventBridge and SNS, the seed of a real detection-and-response pipeline
- closing the loop safely: archiving sample findings and suspending or disabling GuardDuty so no charges accrue, on an account you own
Steps in this lab
- Read this first: your own account, sample findings only, turn it off when done
- What threat detection is -- findings, severity, and where GuardDuty's signal comes from
- Sign in to a free-tier AWS account (and secure the root login)
- Understand the Region picker before you enable anything
- Enable GuardDuty
- Generate sample findings (safe, fake alerts to practice on)
- Read one finding: type, affected resource, and the raw JSON
- Triage by severity -- decide what to deal with first
- Route a finding to a human: send alerts to your email
- Clean up: archive samples, then suspend or disable GuardDuty
- Why this loop makes you a cloud defender