Executive Summary
In 2026, Huntress Labs researchers uncovered a new generation of phishing kits—EvilTokens, Knight Office, and Kali365—engineered specifically to steal and replay Microsoft 365 session tokens, bypassing MFA entirely. These kits operate as full Phishing-as-a-Service platforms with admin dashboards, AI-assisted lure generation, and real-time token relay infrastructure. Using Adversary-in-the-Middle (AiTM) and device code phishing techniques, threat actors capture authenticated session tokens and immediately replay them into live M365 sessions, gaining persistent access to cloud resources, email, and downstream SaaS applications without ever needing the victim's password.
This campaign marks a decisive shift in the credential theft landscape: attackers no longer target passwords—they target identity artifacts. As MFA adoption increases, session hijacking and token replay have become the preferred bypass for enterprise cloud environments, underscoring that completed MFA no longer guarantees session integrity.
Why This Matters Now
MFA bypass via session token theft is now commoditized. Phishing kits like Kali365 and Knight Office are sold as turnkey services, enabling low-skill attackers to compromise M365 tenants at scale in 2026. With Microsoft 365 underpinning collaboration for hundreds of millions of users, the attack surface is vast and the window between token theft and account takeover is measured in seconds—not hours.
Attack Path Analysis
Attackers deployed modern phishing kits (EvilTokens, Knight Office, Kali365) to target Microsoft 365 users via device code phishing, adversary-in-the-middle (AiTM) proxies, and token replay techniques, bypassing MFA entirely by stealing post-authentication session tokens rather than credentials. Once valid session tokens were captured and replayed into live Microsoft 365 sessions, attackers registered rogue devices, established persistence across tenant accounts, accessed mailboxes, and exfiltrated data — all while operating from cloud infrastructure that evaded traditional perimeter controls.
Kill Chain Progression
This analysis maps confirmed threat intelligence to the full cloud kill chain to show where defensive gaps would emerge as an attack progresses.
Initial Compromise
Description
Victims were lured via phishing emails to attacker-controlled pages or device code authentication flows. Phishing kits like Kali365 and Knight Office acted as reverse proxies or PhaaS platforms, intercepting the authentication flow in real time. Victims believed they were completing a legitimate Microsoft 365 sign-in or MFA prompt, while the kit silently harvested the resulting session token. No credential entry on a spoofed page was required in device code flows, making detection harder.
MITRE ATT&CK® Techniques
Phishing: Spearphishing Link
Multi-Factor Authentication Interception
Steal Web Session Cookie
Steal Application Access Token
Use Alternate Authentication Material: Web Session Cookie
Account Manipulation: Device Registration
Email Collection: Remote Email Collection
Adversary-in-the-Middle
Potential Compliance Exposure
Mapping incident impact across multiple compliance frameworks.
PCI DSS 4.0 – Multi-Factor Authentication for Access to the Cardholder Data Environment
Control ID: Requirement 8.4
NYDFS 23 NYCRR 500 – Multi-Factor Authentication
Control ID: Section 500.12
CISA Zero Trust Maturity Model 2.0 – Continuous Identity Validation and Session Risk Evaluation
Control ID: Identity Pillar – Advanced Stage: Continuous Validation
NIS2 Directive – Incident Handling and Access Control Measures
Control ID: Article 21(2)(b) and 21(2)(e)
DORA (Digital Operational Resilience Act) – ICT Access Controls and Incident Classification
Control ID: Article 9 – ICT Security Policies; Article 17 – ICT-Related Incident Management
ISO/IEC 27001:2022 – Secure Authentication Controls and Privileged Session Management
Control ID: Annex A 8.5 – Secure Authentication; Annex A 8.2 – Privileged Access Rights
Sector Implications
Industry-specific impact of the vulnerabilities, including operational, regulatory, and cloud security risks.
Financial Services
Microsoft 365 session token theft via AiTM and device code phishing exposes financial firms to credential compromise, unauthorized cloud access, and PCI/NIST compliance violations requiring zero trust segmentation and egress controls.
Health Care / Life Sciences
Phishing kits stealing M365 session tokens bypass MFA, threatening HIPAA-regulated patient data through lateral movement and exfiltration, demanding token revocation protocols and multicloud visibility controls per HIPAA 164.312 mandates.
Information Technology/IT
IT organizations managing M365 tenants face persistent attacker access via token replay and PhaaS kits like Kali365, requiring inline IPS, cloud firewall egress enforcement, and anomaly detection to contain post-authentication compromise.
Banking/Mortgage
Banking institutions using Microsoft 365 are prime targets for credential and session theft, with stolen refresh tokens enabling unauthorized tenant persistence, threatening compliance with NIST 800-53 SC-7 and PCI 4.0 egress security mandates.
Sources
- Inside Modern Phishing Kits Built to Steal Microsoft 365 Sessionshttps://www.bleepingcomputer.com/news/security/inside-modern-phishing-kits-built-to-steal-microsoft-365-sessions/Verified
- Unboxing Phishing Kits - Huntress Labs Researchhttps://www.huntress.com/blog/unboxing-phishing-kitsVerified
- EvilTokens: AI-Powered Phishing Report - Huntress Labshttps://www.huntress.com/resources/eviltokens-ai-powered-phishing-reportVerified
- Inside Knight Office M365 AiTM Attack - Huntress Labshttps://www.huntress.com/blog/inside-knight-office-m365-aitm-attackVerified
- Kali365 Device Code Phishing Kit - Huntress Labshttps://www.huntress.com/blog/kali365-device-code-phishing-kitVerified
- Tradecraft Tuesday: Device Code Phishing Explained - Huntress Labshttps://www.huntress.com/blog/tradecraft-tuesday-device-code-phishing-explainedVerified
- CISA Known Exploited Vulnerabilities Cataloghttps://www.cisa.gov/known-exploited-vulnerabilities-catalogVerified
- Microsoft Identity Platform - OAuth 2.0 Device Authorization Grant Flow Security Guidancehttps://learn.microsoft.com/en-us/entra/identity-platform/v2-oauth2-device-codeVerified
- Phishing Resistant MFA and Conditional Access - Microsoft Security Bloghttps://www.microsoft.com/en-us/security/blog/2022/09/22/token-tactics-how-to-prevent-detect-and-respond-to-cloud-token-theft/Verified
Frequently Asked Questions
Cloud Native Security Fabric Mitigations and ControlsCNSF
Based on the attack progression modeled above, these are the defensive controls that would constrain each stage.
This incident demonstrates how token-based identity attacks can bypass perimeter controls entirely, making workload-level segmentation and egress enforcement critical layers of residual defense. Aviatrix Zero Trust CNSF would likely reduce attacker reachability across cloud workloads and constrain the blast radius even after valid session tokens were captured and replayed.
Control: Cloud Native Security Fabric (CNSF)
Mitigation: Outbound access to attacker-controlled phishing infrastructure operating from cloud PaaS environments would likely be constrained, reducing the number of successful token interception opportunities originating from managed cloud workloads.
Control: Zero Trust Segmentation
Mitigation: Zero Trust segmentation would likely constrain the scope of what an attacker operating with a replayed token could reach within the cloud environment, reducing lateral privilege scope even when identity artifacts appear fully authenticated.
Control: East-West Traffic Security
Mitigation: East-west traffic enforcement would likely constrain an attacker's ability to move freely between cloud workloads and services after gaining initial access, reducing reachability across internal segmentation boundaries even with a valid session token in hand.
Control: Multicloud Visibility & Control
Mitigation: Multicloud visibility and control would likely surface anomalous outbound connection patterns toward attacker-operated cloud infrastructure, reducing the ability of token refresh and command relay traffic to persist undetected across managed cloud environments.
Control: Egress Security & Policy Enforcement
Mitigation: Egress security policy enforcement would likely reduce the volume of data exfiltrated by constraining outbound paths to only approved destinations, limiting how much harvested credential and session data could reach attacker-controlled endpoints through unmanaged cloud egress channels.
Even if attacker persistence was established through rogue device registration and token replay, the constrained segmentation and egress posture enforced by prior CNSF controls would likely have reduced the scope of accessible cloud assets, limiting downstream BEC impact and the volume of data exposed to ongoing unauthorized access.
Impact at a Glance
Affected Business Functions
- Microsoft 365 Cloud Email and Collaboration Services
- Identity and Access Management (IAM) / Single Sign-On (SSO)
- Corporate Data and Document Repositories (SharePoint, OneDrive)
- Security Operations and Incident Response
Estimated downtime: 3 days
Estimated loss: N/A
Victims of session token theft targeting Microsoft 365 environments are at risk of exposure across multiple sensitive data categories: corporate email communications including confidential business correspondence, financial reports, and HR records; cloud-stored documents and files in SharePoint and OneDrive; internal credentials and authentication artifacts enabling lateral movement to downstream SaaS platforms; identity artifacts such as registered device credentials and Windows Hello keys enabling persistent re-entry; and potential exposure of customer PII, proprietary business data, and regulated information depending on the victim organization's industry. Because token replay grants access equivalent to a fully authenticated MFA-verified session, the attacker gains the same data access as the legitimate user with no further authentication barriers. The scale of exposure is amplified by the PhaaS model of kits like Kali365, which enables high-volume campaigns targeting organizations across multiple sectors simultaneously.
Recommended Actions
Key Takeaways & Next Steps
- • Apply Cloud Native Security Fabric (CNSF) inline enforcement and real-time traffic inspection to detect anomalous OAuth token issuance, automated token refresh patterns, and suspicious device code authentication flows originating from unfamiliar cloud infrastructure before sessions are established.
- • Deploy Multicloud Visibility & Control capabilities to baseline normal Microsoft 365 authentication behavior and alert on repeated malformed or automated token requests, reverse-proxy relay patterns, and authentication originating from PaaS or cloud-hosted attacker infrastructure — enabling early detection of AiTM and device code phishing activity.
- • Enforce Egress Security & Policy Enforcement with FQDN filtering and outbound traffic controls to block data exfiltration paths used by phishing kit operator consoles, prevent unauthorized token relay to attacker-controlled PaaS endpoints, and detect outbound mailbox forwarding rule creation that enables persistent email exfiltration.
- • Implement Zero Trust Segmentation with identity-based policy to enforce least-privilege access within Microsoft 365 tenants — restricting lateral movement across services (Exchange, SharePoint, Teams) even after a valid session token is replayed, and limiting the blast radius of any compromised identity artifact.
- • Use Cloud Firewall (ACF) with AI-driven traffic discovery and internet egress controls to identify and block traffic to known phishing kit infrastructure, unauthorized reverse proxy endpoints, and Railway PaaS or similar platforms used as token-harvesting engines, while continuously monitoring for new unauthorized outbound destinations associated with phishing-as-a-service operations.



