Detections
Explore public detection logic contributed by the community across SIEM and rule languages.
3,866 detections
Filters
Last updated
All Time
Detection languages
2,525
525
299
218
131
Contributors
999
706
554
263
188
Categories
1,449
1,199
1,091
792
575
Platforms
39,936
7,266
6,537
3,866
3,647
Products / Services
1,755
1,449
1,121
438
406
MITRE Techniques
1,889
1,392
913
814
727
CVEs
12
5
4
3
2
IDS Classtypes
10
7
6
6
5
IDS Protocols
37
2
1
Detects tenant-level enablement of Microsoft Teams external/federated access or addition of an unfamiliar external domain shortly before a spike in inbound Teams calls to helpdesk/IT-support-tagged users — models CrowdStrike's 2026 Financial Services Threat Landscape Report on 'Mutant Spider,' the most active threat to financial services, whose primary technique is voice phishing over Microsoft Teams impersonating internal IT.
This rule detects scenarios where a user account modifies Conditional Access (CA) policies (specifically targeting user or group inclusion/exclusion settings) within one hour of that same user account modifying authentication method policies. This pattern is potentially indicative of an adversary weakening security controls by adjusting authentication requirements and subsequently tampering with CA policies to maintain persistence or bypass MFA.
This rule detects internal users submitting potentially sensitive data to public generative AI applications (e.g., ChatGPT, Claude, Gemini) where Microsoft Purview DLP or a comparable inspection service has triggered a policy match. It correlates cloud application outbound signals with content inspection logs to identify high-risk exfiltration events while excluding sanctioned enterprise-licensed AI instances.
Single-source Sigma proxy for the authentication-persistence stage of the multi-stage device-code-phishing-to-Graph-collection chain (see the companion KQL correlation rule "Passkey Persistence: Device-Code Phishing to MFA Takeover and Graph Collection" for the full, threshold-gated, cross-table detection). Sigma has no join, temporal-windowing, or aggregation support, so this rule fires on the authentication-persistence event alone and does NOT enforce the device-code sign-in precursor or the subsequent Graph-activity burst - both required by the full correlation. Scoped to the two highest-signal operations only (new security-info registration, and any admin-initiated registration on another user's account) to cut noise from routine self-service profile edits.
Single-source Sigma proxy for the authentication-persistence stage of the multi-stage device-code-phishing-to-Graph-collection chain (see the companion KQL correlation rule "Passkey Persistence: Device-Code Phishing to MFA Takeover and Graph Collection" for the full, threshold-gated, cross-table detection). Sigma has no join, temporal-windowing, or aggregation support, so this rule fires on the authentication-persistence event alone and does NOT enforce the device-code sign-in precursor or the subsequent Graph-activity burst - both required by the full correlation. Scoped to the two highest-signal operations only (new security-info registration, and any admin-initiated registration on another user's account) to cut noise from routine self-service profile edits.
Detects the multi-stage Microsoft cloud identity compromise chain reported by Microsoft Security Research (September 2026): a victim completes a successful Entra ID device-code sign-in (often induced by passkey-themed device-code phishing), the same identity's authentication/security info is registered or changed (MFA/passkey persistence) within 60 minutes (deduplicated to the earliest such event per sign-in), followed within 120 minutes by a burst of at least 25 successful, delegated-context Microsoft Graph API calls (Scopes claim populated, i.e. user-delegated permissions rather than pure application/client-credential access) spanning at least 3 distinct resource paths, exhibiting reconnaissance (>=5 requests against /users, /groups, /directoryRoles, /roleManagement, /applications, /servicePrincipals, /organization) and/or collection (>=10 requests against /messages, /mailFolders, /attachments, /drive, /drives, /sites, /lists, /search) patterns. Restricting to delegated (Scopes-bearing) Graph calls and requiring resource-path breadth removes the two largest false-positive sources: service-principal/application-only automation jobs (Roles-only tokens) and single-endpoint polling tools. Pure behavioral correlation across SigninLogs, AuditLogs, and MicrosoftGraphActivityLogs - no static IOCs. Flags whether Graph activity originated from an IP different from the original device-code sign-in IP as a strong pivot/session-reuse indicator.
Detects instances where a user account registers a new Multi-Factor Authentication (MFA) method (such as TOTP or Authenticator app) shortly (within 30 minutes) after a high-risk or potentially anomalous sign-in event. This pattern is commonly associated with adversary-in-the-middle (AiTM) phishing attacks where an attacker captures an authenticated session and uses it to enroll a secondary, attacker-controlled MFA device for persistent account access.
Detects a single identity listing access keys for a high volume (20+) of Azure Storage Accounts within a 15-minute window. This behavior is indicative of credential harvesting, reconnaissance, or automated exfiltration attempts targeting storage resources.
Detects Azure AD sign-in events where the user-agent matches common scripting or automation libraries such as python-requests. This behavior is often indicative of automated token replay attacks, where attackers attempt to leverage stolen session cookies or tokens (e.g., following AiTM phishing) via proxy infrastructure.
Detects rapid, automated-style creation of an application, associated service principal, and delegated permission grants within a 60-second window. This behavior is indicative of the TrustSink technique or similar automated adversary infrastructure setup in Microsoft Entra ID (formerly Azure AD), where an attacker establishes persistence or elevated access through newly created apps.
Detects AI agent tooling (e.g., LangChain, Copilot, Claude Code) performing read operations on databases containing sensitive data patterns, or performing suspicious bulk data enumeration. The rule monitors Azure SQL Database security audit events for suspicious application name signatures and non-schema introspective queries.
Correlates a successful OAuth device-code sign-in with a subsequent Device Registration Service event for the same user within 30 minutes, excluding known legitimate device-code applications (Azure CLI, Azure PowerShell, Visual Studio, VS Code) and requiring both events to have succeeded. The registration match is scoped strictly to the Device Registration Service resource (not the broader Microsoft Authentication Broker app, which fires on routine token refresh/broker activity unrelated to registering a new device). The correlation is further restricted to cases where the device-registration event originates from a different IP address or city than the original device-code sign-in -- the pattern expected when an attacker uses a stolen token to register their own device and mint a Primary Refresh Token, versus a legitimate user completing device enrollment from their own network shortly after a CLI/developer login.
Detects anomalous external traffic patterns indicative of automated API reconnaissance. The rule flags high-frequency requests from a single IP targeting a diverse set of API endpoints, particularly when characterized by a mixture of successful (200), unauthorized (403), or not found (404) status codes, or repeated attempts to access sensitive management/administrative API paths.
Detects anomalous post-authentication reconnaissance activities against the Microsoft Graph API, specifically the enumeration of directory objects, roles, or service principals. The rule triggers when a user account performs multi-endpoint enumeration via the Graph API within 30 minutes of a successful authentication session initiated via a device code flow that was flagged with elevated risk signals. This behavior is consistent with post-compromise reconnaissance techniques such as 'EvilTokens', where an adversary attempts to map the environment after gaining access.
Detects changes to Entra ID (Azure AD) federation settings, specifically targeting updates to trust and authentication parameters such as IssuerUri, ActiveLogOnUri, and TokenSigningCertificate. These modifications are critical indicators of potential Golden SAML attacks, where adversaries forge SAML tokens to bypass MFA and impersonate users.
Detects when high-privilege application permissions (e.g., Directory.ReadWrite.All, Application.ReadWrite.All) are granted to a service principal via admin consent. This behavior can be used by attackers to gain persistent, stealthy administrative access to a Microsoft Entra ID tenant.
Detects modifications to Entra ID (Azure AD) cross-tenant access settings, including inbound/outbound trust settings and B2B collaboration configurations. This activity may indicate an adversary attempting to establish persistence or facilitate unauthorized access by creating a trust relationship with an external tenant they control.
Detects modifications to Entra ID (Azure AD) cross-tenant access settings, including inbound/outbound trust settings and B2B collaboration configurations. This activity may indicate an adversary attempting to establish persistence or facilitate unauthorized access by creating a trust relationship with an external tenant they control.
This rule detects when a user or admin grants consent to an unverified OAuth application that simultaneously requests sensitive delegated permissions across multiple categories (Mail, Files, Directory, and Contacts). This pattern is a strong indicator of consent phishing attacks, where adversaries trick users into granting broad permissions to a malicious application to harvest sensitive data or gain persistent access to a Microsoft 365 environment.
Detects anomalous Privileged Identity Management (PIM) role activations in Entra ID that deviate from standard patterns. This includes instances where roles are activated without required approvals, for unusually long durations, or when an account activates multiple distinct roles within a short time frame, potentially indicating account takeover or abuse of standing privileges.
Detects non-interactive Microsoft Entra ID service principal or managed identity sign-in events originating from an IP address, ASN, or country that has not been historically associated with that specific service principal. This behavior may indicate the use of compromised service principal credentials or stolen access tokens from adversary-controlled infrastructure.
Page 24 of 185