Revoke the session, inspect mailbox rules and recent sends, reset the account, and check whether the same infrastructure touched other users. Then review whether the account was allowed to operate from unmanaged devices, because token replay is far easier when session controls are weak.
What a suspected Microsoft 365 token theft means operationally
A stolen Microsoft 365 token should be treated as active authenticated access, not just a password problem. The immediate question is whether the token is still usable, whether the session has already been abused, and whether the account’s permissions let the attacker reach mail, files, or connected apps. That makes containment, trace review, and session invalidation the priority.
Because token replay can preserve access even after a password reset, teams should assume the attacker may be working inside normal sign-in patterns until the token and any linked sessions are invalidated. That is why mailbox rules, recent message activity, and other persistence checks matter as much as the reset itself.
What to check first in Microsoft 365
Start by revoking the active session and any refresh paths that can mint new access. Then inspect mailbox rules, forwarding settings, sent items, delegated access, and recent OAuth consent or app activity. If the account touched shared mailboxes, OneDrive, or Teams, expand the review to those surfaces before declaring the incident contained.
Look for signs that the token was used from unfamiliar infrastructure or at unusual times, and compare that activity with other users who may have been exposed through the same campaign. When the same access path appears across multiple accounts, the issue is often broader than a single mailbox compromise.
How to contain the account and prevent replay
Revoke sessions, reset the account, and reissue credentials or secrets only after you understand which access paths were actually exposed. If the tenant allows unmanaged devices or weak session controls, tighten those conditions quickly because they are common enablers of token replay and silent persistence.
For Microsoft 365 specifically, teams should also review whether the compromised identity had elevated privileges, consented applications, or mailbox delegation that would let the attacker move laterally without triggering obvious password-based alerts. A token compromise is often an authorization problem as much as an authentication problem.
Risk and Threat Considerations
A stolen Microsoft 365 token can remain usable until the session is revoked, so the main risk is not just mailbox access but continued replay across email, files, and connected SaaS apps. The attacker may use the valid session to hide in normal traffic, create mailbox rules, forward data out, or pivot through trusted integrations.
Failure mechanism: The token bypasses fresh password prompts, and weak session governance lets the attacker keep operating from an already trusted browser, device, or app context.
Impact: Teams can miss the compromise window, lose control of email and collaboration data, and face repeated re-entry even after a password reset if related sessions or consented apps are left intact.
Standards & Framework Alignment
This section maps relevant standards and security frameworks to the operational risks and controls described in this guidance.
NIST SP 800-53 Rev 5, CIS Controls v8 and OWASP ASVS set the governance and control requirements practitioners need to meet.
| Framework | Control / Reference | Relevance |
|---|---|---|
| NIST SP 800-53 Rev 5 | IA-5 — Authenticator Management | Token theft response depends on revoking and reissuing authenticators. |
| AC-2 — Account Management | Account reset and access review are central after suspected token theft. | |
| AU-6 — Audit Record Review, Analysis, and Reporting | Mailbox and sign-in review relies on audit evidence to confirm abuse. | |
| Recommendation — Revoke compromised authenticators and replace them before restoring access. Review account state and disable or re-enable access only after containment. Analyze sign-in and mailbox audit records for signs of token replay and persistence. | ||
| CIS Controls v8 | CIS-5 — Account Management | Suspicious token use calls for rapid account and session control. |
| Recommendation — Remove stale access and verify account ownership before restoring use. | ||
| OWASP ASVS | V7 — Session Management | The core issue is whether an existing session can still be replayed. |
| Recommendation — Invalidate compromised sessions and enforce stronger session controls. | ||
Practitioner Guidance
What to prioritise: Treat session invalidation and persistence hunting as the first response, not the last. If you only reset the account, you may leave the attacker’s working session alive long enough to exfiltrate data or expand access.
What to verify: Confirm that the token is no longer usable, mailbox rules were not altered, recent sends are explainable, and no connected app or delegated path still authorises the same account. If unmanaged device access was allowed, verify whether that gap contributed to the replay path.
Practitioner takeaway: A suspected token theft is a live access incident, so response should focus on shutting down the session, checking for persistence, and determining whether the attacker can still re-enter through the same trust path.
Related resources from NHI Mgmt Group
- How should security teams respond when a SaaS session token is stolen?
- How should security teams respond when a Microsoft 365 AiTM phishing kit is delivering a live session relay instead of just stealing passwords?
- How should security teams respond when a third-party integration token is stolen and starts accessing cloud data stores?
- How should security teams investigate a suspected Microsoft 365 BEC lead?
Deepen Your Knowledge
Free weekly newsletter
Subscribe to the NHI & AI Identity Journal
The latest on NHI and Agentic AI security – articles, research, breaches, news and events every week.
Bonus 33% off our NHI Course when you subscribe.
Reviewed and updated by the NHIMG editorial team on October 23, 2026.
NHI Mgmt Group — the #1 independent authority on Non-Human Identity, IAM, and Agentic AI security. nhimg.org