- From
- Dana Park <dana.park@northwind-demo.example>
- To
- soc@northwind-demo.example
- Date
- 2026-04-19 09:42 UTC
Finance user entered credentials on a fake Microsoft 365 login
# Web proxy (src=10.20.4.51 dana-laptop) 09:24:02 GET https://m365-northwind-login[.]example/auth 200 09:24:38 POST https://m365-northwind-login[.]example/auth/submit 302 09:24:39 GET https://office.com/ 200 # Identity provider sign-in log (user: dana.park@northwind-demo.example) 09:25:10 SUCCESS ip=203.0.113.77 (hosting provider) ua="python-requests/2.31" mfa=not_challenged 09:25:48 SUCCESS ip=203.0.113.77 app="Outlook Web" action=New-InboxRule "move-to-archive"
- Name
- dana.park@northwind-demo.example
- Type
- Finance user account + laptop (dana-laptop)
- Owner
- Finance Dept · Dana Park
- Level
- High
First, I escalate this as a P1 critical incident, because a Finance user was phished and a malicious sign-in has already succeeded. The replayed session with no MFA challenge tells me this is adversary-in-the-middle session token theft, not just a stolen password. This social engineering attack impersonated a trusted internal sender. For containment, I reset password and rotate credentials on the affected account, revoke session tokens and sign out everywhere so the stolen refresh token is invalidated, and disable account access if suspicious activity continues. I isolate the laptop and block the phishing domain and the malicious source IP at the proxy. Next, for investigation I review the identity provider sign-in log, confirm the source IP, geolocation, and user agent, and audit the mailbox for the attacker-created inbox rule or any auto-forward. I gather indicators of compromise (IOC) and check whether other users were targeted by the same page. Before remediating, I preserve evidence: I export logs, capture a forensic snapshot of the laptop, and retain the message with its original headers for chain of custody. Finally, for recovery I restore access once the account is verified clean, enforce MFA through Conditional Access, and run phishing awareness training so a repeat is less likely.
Solid response — your plan covers the core incident response steps and avoids dangerous actions. Score: 97/100. Strongest area: Clarity & structure (100%). Weakest area: Asset impact (67%) — expand this next time.
Where points came from
- Attack understanding3/3 · 15.0 / 15
- Asset impact2/3 · 6.7 / 10
- Prioritization2/2 · 10.0 / 10
- Containment5/5 · 20.0 / 20
- Investigation4/4 · 15.0 / 15
- Recovery3/3 · 10.0 / 10
- Evidence preservation3/3 · 10.0 / 10
- Clarity & structure2/2 · 10.0 / 10
Strengths
- Attack understanding
- Prioritization
- Containment
- Investigation
- Recovery
- Evidence preservation
- Clarity & structure
Missing / weak
No category dropped below 40%.
Dangerous actions detected
None detected in your response.
Sample AI coaching
This is illustrative sample coaching for the demo, not live AI output. A strong response like the one above leads with prioritization and containment before investigation, and it preserves evidence before remediating. To go further, tie each action to the specific evidence — the replayed session from the hosting-provider IP and the attacker-created inbox rule — and state exactly who you would notify and when. On the live trainer this section can be produced by the optional AI Review layer; here it is fixed sample text.