AWS

Detection rules for AWS data sources. Each rule includes its MITRE ATT&CK mapping, impact rating, and a description of what it detects. These rules analyze events with the data type aws, which the AWS CloudWatch integration sends.

This category contains 81 detection rules.

RuleCategoryTechniqueImpact (C/I/A)
AWS Account Closure RequestedImpactT1485 - Data DestructionC:3 / I:3 / A:3
AWS Backup Deletion for ImpactImpactT1490 - Inhibit System RecoveryC:1 / I:3 / A:3
AWS CloudFormation Stack DeletionImpactT1485 - Data DestructionC:1 / I:3 / A:3
AWS CloudTrail Event Selector ManipulationDefense EvasionT1562.008 - Impair Defenses: Disable Cloud LogsC:3 / I:3 / A:2
AWS CloudTrail Log DeletedDefense EvasionT1562 - Impair DefensesC:2 / I:3 / A:2
AWS CloudTrail Log SuspendedDefense EvasionT1562 - Impair DefensesC:3 / I:2 / A:2
AWS CloudTrail Log UpdatedImpactT1565 - Data ManipulationC:2 / I:2 / A:1
AWS CloudTrail Logging DisabledDefense EvasionT1562.008 - Impair Defenses: Disable Cloud LogsC:3 / I:3 / A:2
AWS CloudTrail Trail CreatedDefense EvasionT1562 - Impair DefensesC:3 / I:2 / A:1
AWS CloudWatch Log Group DeletionImpactT1485 - Data DestructionC:3 / I:2 / A:2
AWS CloudWatch Log Stream DeletionImpactT1485 - Data DestructionC:3 / I:2 / A:2
AWS Compute-Intensive Instance Launch RequestResource HijackingT1496 - Resource HijackingC:1 / I:2 / A:3
AWS Config Service DisabledDefense EvasionT1562.008 - Impair Defenses: Disable Cloud LogsC:2 / I:3 / A:2
AWS Config Service TamperingDefense EvasionT1562 - Impair DefensesC:3 / I:2 / A:2
AWS Configuration Recorder StoppedDefense EvasionT1562 - Impair DefensesC:3 / I:2 / A:2
AWS Console GetSigninToken AbuseCredential AccessT1550.001 - Use Alternate Authentication Material: Application Access TokenC:3 / I:2 / A:1
AWS Console Logins from Different CountriesCredential AccessT1078.004 - Valid Accounts: Cloud AccountsC:3 / I:2 / A:1
AWS Cross-Account Access AnomaliesUnauthorized AccessT1550.001 - Use Alternate Authentication Material: Application Access TokenC:3 / I:3 / A:1
AWS EBS Snapshot Sharing Change for ReviewExfiltrationT1537 - Transfer Data to Cloud AccountC:3 / I:2 / A:1
AWS EBS Snapshot Sharing Permission AddedExfiltrationT1537 - Transfer Data to Cloud AccountC:3 / I:2 / A:1
AWS EC2 Encryption DisabledImpactT1565 - Data ManipulationC:3 / I:2 / A:2
AWS EC2 Flow Log DeletionDefense EvasionT1562 - Impair DefensesC:3 / I:2 / A:2
AWS EC2 Instance Startup Script ModifiedExecutionT1059 - Command and Scripting InterpreterC:3 / I:3 / A:2
AWS EC2 Metadata Token Requirement RelaxedCredential AccessT1552.005 - Unsecured Credentials: Cloud Instance Metadata APIC:3 / I:2 / A:1
AWS EC2 Snapshot ActivityExfiltrationT1537 - Transfer Data to Cloud AccountC:3 / I:2 / A:2
AWS EC2 Traffic Mirroring Configuration RequestExfiltrationT1020 - Automated ExfiltrationC:3 / I:2 / A:2
AWS EC2 VM Export RequestExfiltrationT1537 - Transfer Data to Cloud AccountC:3 / I:2 / A:2
AWS ECS/EKS Container Abuse DetectionExecutionT1610 - Deploy ContainerC:3 / I:3 / A:2
AWS Glue Development Endpoint Privilege EscalationPrivilege EscalationT1078.004 - Valid Accounts: Cloud AccountsC:3 / I:3 / A:2
AWS GuardDuty Detector DeletionDefense EvasionT1562 - Impair DefensesC:3 / I:2 / A:2
AWS GuardDuty Detector Updated or DisabledDefense EvasionT1685 - Disable or Modify Cloud LogsC:3 / I:2 / A:1
AWS GuardDuty Finding Filter AddedDefense EvasionT1685 - Disable or Modify Cloud LogsC:3 / I:2 / A:1
AWS GuardDuty High or Critical FindingThreat DetectionVendor threat detectionC:3 / I:3 / A:2
AWS IAM Assume Role Policy UpdateInitial AccessT1078 - Valid AccountsC:2 / I:2 / A:1
AWS IAM Backdoor Creation AttemptsPrivilege EscalationT1136.003 - Create Account: Cloud AccountC:3 / I:3 / A:2
AWS IAM Deactivation of MFA DeviceImpactT1531 - Account Access RemovalC:3 / I:2 / A:2
AWS IAM Login Profile Modified for Another UserPersistenceT1098 - Account ManipulationC:3 / I:3 / A:1
AWS IAM Password Recovery RequestedInitial AccessT1078 - Valid AccountsC:2 / I:1 / A:0
AWS IAM Privilege Escalation Path DetectionPrivilege EscalationT1098 - Account ManipulationC:3 / I:3 / A:2
AWS KMS Key Disabled or Deletion ScheduledImpactT1486 - Data Encrypted for ImpactC:3 / I:3 / A:1
AWS KMS Key Material Import or DeletionSystem IntegrityT1486 - Data Encrypted for ImpactC:3 / I:3 / A:3
AWS Lambda Function URL Backdoor CreationPersistenceT1059 - Command and Scripting InterpreterC:3 / I:3 / A:1
AWS Management Console Brute Force of Root User IdentityCredential AccessT1110 - Brute ForceC:3 / I:2 / A:1
AWS Management Console Root LoginInitial AccessT1078 - Valid AccountsC:3 / I:3 / A:3
AWS Mass Resource DeletionImpactT1485 - Data DestructionC:1 / I:3 / A:3
AWS RDS Cluster DeletionImpactT1485 - Data DestructionC:3 / I:2 / A:2
AWS RDS Deletion Protection Disabled or Master Password ChangedImpactT1485 - Data DestructionC:3 / I:3 / A:1
AWS RDS Publicly Accessible Restore RequestedSystem IntegrityT1537 - Transfer Data to Cloud AccountC:3 / I:2 / A:1
AWS RDS Security Group Internet Ingress AddedDefense EvasionT1562.007 - Impair Defenses: Disable or Modify Cloud FirewallC:3 / I:2 / A:2
AWS RDS Snapshot Export Request AcceptedExfiltrationT1537 - Transfer Data to Cloud AccountC:3 / I:2 / A:2
AWS RDS Snapshot Sharing for Data ExfiltrationData ExfiltrationT1537 - Transfer Data to Cloud AccountC:3 / I:2 / A:1
AWS Root Account Usage Without MFADefense Evasion, Persistence, Privilege Escalation, Initial AccessT1078.004 - Valid Accounts: Cloud AccountsC:3 / I:3 / A:2
AWS Root Login Without MFAInitial AccessT1078 - Valid AccountsC:3 / I:2 / A:2
AWS Route 53 DNS Hijacking AttemptInitial AccessT1584.002 - Compromise Infrastructure: DNS ServerC:3 / I:3 / A:3
AWS Route 53 Domain Transfer Lock DisabledPersistenceT1098 - Account ManipulationC:3 / I:2 / A:2
AWS Route 53 Domain Transfer Request AcceptedPersistenceT1098 - Account ManipulationC:3 / I:3 / A:3
AWS S3 Bucket Configuration DeletionDefense EvasionT1070 - Indicator RemovalC:3 / I:2 / A:2
AWS S3 Bucket Server Access Logging DisabledDefense EvasionT1562.008 - Disable Cloud LogsC:3 / I:2 / A:1
AWS S3 Bucket Versioning SuspendedImpactT1490 - Inhibit System RecoveryC:2 / I:3 / A:3
AWS S3 Bulk Data Exfiltration DetectedData ExfiltrationT1530 - Data from Cloud Storage ObjectC:2 / I:1 / A:1
AWS S3 Public Access Controls RelaxedCollectionT1530 - Data from Cloud Storage ObjectC:3 / I:2 / A:1
AWS SAML Role Assumption After Provider ChangeCredential AccessT1556 - Modify Authentication ProcessC:3 / I:3 / A:3
AWS SSM SendCommand Remote ExecutionExecutionT1059 - Command and Scripting InterpreterC:2 / I:3 / A:2
AWS SSO Identity Provider Configuration ChangedCredential AccessT1556 - Modify Authentication ProcessC:3 / I:3 / A:2
AWS STS Token Abuse DetectionPrivilege EscalationT1078.004 - Cloud AccountsC:3 / I:3 / A:1
AWS Secrets Manager Suspicious Access PatternCredential AccessT1555.006 - Credentials from Password Stores: Cloud Secrets Management StoresC:3 / I:2 / A:1
AWS Security Group ModificationsDefense EvasionT1562.007 - Impair Defenses: Disable or Modify Cloud FirewallC:2 / I:3 / A:2
AWS Security Hub DisabledDefense EvasionT1685 - Disable or Modify Cloud LogsC:3 / I:2 / A:1
AWS SecurityHub Finding EvasionDefense EvasionT1562 - Impair DefensesC:2 / I:3 / A:2
AWS Systems Manager Command Request AcceptedInitial AccessT1566 - PhishingC:2 / I:1 / A:1
AWS Systems Manager Session AbuseLateral MovementT1021 - Remote ServicesC:3 / I:3 / A:2
AWS TruffleHog Credential Scanning DetectedCredential AccessT1552 - Unsecured CredentialsC:3 / I:1 / A:0
AWS Unusual API Call PatternsDiscoveryT1580 - Cloud Infrastructure DiscoveryC:2 / I:1 / A:1
AWS WAF Access Control List DeletionDefense EvasionT1562 - Impair DefensesC:3 / I:2 / A:2
AWS WAF Rule or Rule Group DeletionDefense EvasionT1562 - Impair DefensesC:3 / I:2 / A:2
AWS WAF and Shield Rule ModificationsDefense EvasionT1562.001 - Impair Defenses: Disable or Modify ToolsC:2 / I:3 / A:3
Repeated AWS Administrative Policy AttachmentsPrivilege EscalationT1098 - Account ManipulationC:3 / I:3 / A:2
Repeated AWS SSO Permission Set ChangesPrivilege EscalationT1098 - Account ManipulationC:3 / I:3 / A:1
Repeated AWS VPC Flow Log Deletion RequestsDefense EvasionT1562 - Impair DefensesC:3 / I:2 / A:2
Repeated ECS Task API Requests by an Assumed RoleSystem IntegrityT1610 - Deploy ContainerC:3 / I:2 / A:1
Repeated Malformed AWS IAM Trust Policy UpdatesPrivilege EscalationT1098 - Account ManipulationC:3 / I:2 / A:1

Rule Example

Below is an example of a rule definition for AWS Account Closure Requested (view in repository):

# Rule contract revision 2026-09-25; SDK v1.1.36 (gap-fill set).
dataTypes:
- aws
name: AWS Account Closure Requested
impact:
  confidentiality: 3
  integrity: 3
  availability: 3
category: Impact
technique: T1485 - Data Destruction
adversary: origin
description: Identifies a CloseAccount request, the irreversible end-state of account compromise or destructive misuse.
references:
- https://docs.aws.amazon.com/accounts/latest/reference/close-account.html
where: |
  equals("log.awsRecordType","cloudtrail") &&
  (equals("log.eventSource", "account.amazonaws.com") &&
  equals("log.eventName", "CloseAccount") &&
  equals("actionResult","success"))
groupBy:
- lastEvent.dataSource
- lastEvent.log.awsAccountKeyType
- lastEvent.log.awsAccountKey
- adversary.ip
- adversary.user

Rule Details

AWS Account Closure Requested

Identifies a CloseAccount request, the irreversible end-state of account compromise or destructive misuse.

AWS Backup Deletion for Impact

Detects deletion of AWS Backup vaults, recovery points, or backup plans which could indicate an attacker attempting to prevent recovery after a destructive attack. Removing backups is a common precursor to ransomware or data destruction campaigns.

AWS CloudFormation Stack Deletion

Detects deletion of CloudFormation stacks which could indicate destructive actions by an attacker or unauthorized infrastructure changes. This rule monitors for DeleteStack operations that could result in loss of critical infrastructure. Multiple stack deletions from the same user within a short time frame may indicate malicious activity.

AWS CloudTrail Event Selector Manipulation

Detects manipulation of CloudTrail event selectors which can be used to selectively exclude data events from logging. Attackers may use PutEventSelectors to disable logging of S3, Lambda, or DynamoDB data events to hide their data exfiltration or manipulation activities.

AWS CloudTrail Log Deleted

Identifies the deletion of an AWS log trail. An adversary may delete trails in an attempt to evade defenses

AWS CloudTrail Log Suspended

Identifies suspending the recording of AWS API calls and log file delivery for the specified trail. An adversary may suspend trails in an attempt to evade defenses

AWS CloudTrail Log Updated

Identifies an update to an AWS log trail setting that specifies the delivery of log files

AWS CloudTrail Logging Disabled

Detects attempts to disable CloudTrail logging which could be used to hide malicious activities and evade detection. CloudTrail provides audit logs of all AWS API calls and is critical for security monitoring and compliance.

AWS CloudTrail Trail Created

Identifies creation of a new CloudTrail trail, which an attacker may use to restrict which events are recorded or to ship logs to an adversary-controlled bucket.

AWS CloudWatch Log Group Deletion

Identifies the deletion of a specified AWS CloudWatch log group. When a log group is deleted, all the archived log events associated with the log group are also permanently deleted

AWS CloudWatch Log Stream Deletion

Identifies the deletion of an AWS CloudWatch log stream, which permanently deletes all associated archived log events with the stream

AWS Compute-Intensive Instance Launch Request

Detects a successful RunInstances request for the listed GPU or large compute instance types. These types have legitimate uses; verify authorization, requested capacity and workload. The instance type does not establish cryptomining.

AWS Config Service Disabled

Detects attempts to disable AWS Config by stopping the configuration recorder or deleting the delivery channel. AWS Config tracks resource configuration changes and compliance, and disabling it is a common defense evasion technique.

AWS Config Service Tampering

Identifies attempts to delete an AWS Config Service resource. An adversary may tamper with Config services in order to reduce visibility into the security posture of an account and / or its workload instances

AWS Configuration Recorder Stopped

Identifies an AWS configuration change to stop recording a designated set of resources

AWS Console GetSigninToken Abuse

Detects GetSigninToken API calls which allow federated users to get a console sign-in token. Attackers abuse this to pivot from programmatic credentials to console access, potentially bypassing MFA requirements.

AWS Console Logins from Different Countries

Detects successful console logins by the same principal and account from different geolocated countries within 30 minutes. VPNs, proxies and geolocation changes can produce this sequence; it does not establish impossible travel. Verify the sign-ins with the account owner.

AWS Cross-Account Access Anomalies

Detects anomalous cross-account access patterns in AWS that may indicate account compromise or privilege escalation. Monitors for AssumeRole activities across different accounts where the assumed role ARN does not match the originating account ID, potentially indicating unauthorized cross-account access.

AWS EBS Snapshot Sharing Change for Review

Detects a successful addition of EBS snapshot volume-creation permissions. Check whether the destination accounts or public group are authorized. Removing permissions does not match, and a sharing change alone does not establish a policy violation.

AWS EBS Snapshot Sharing Permission Added

Detects a successful ModifySnapshotAttribute request that adds volume-creation permissions. Removal-only changes do not match. Review the added accounts or public group; sharing permission does not establish completed data exfiltration.

AWS EC2 Encryption Disabled

Identifies disabling of Amazon Elastic Block Store (EBS) encryption by default in the current region. Disabling encryption by default does not change the encryption status of your existing volumes

AWS EC2 Flow Log Deletion

Identifies the deletion of one or more flow logs in AWS Elastic Compute Cloud (EC2). An adversary may delete flow logs in an attempt to evade defenses

AWS EC2 Instance Startup Script Modified

Detects ModifyInstanceAttribute API calls targeting userData, which modifies the EC2 instance startup script. Attackers use this for persistent code execution with root/SYSTEM privileges on instance start.

AWS EC2 Metadata Token Requirement Relaxed

Detects a successful ModifyInstanceMetadataOptions request setting httpTokens to optional. This permits use of IMDSv1; it does not prove that instance credentials were accessed. Verify the change and instance metadata configuration.

AWS EC2 Snapshot Activity

An attempt was made to modify AWS EC2 snapshot attributes. Snapshots are sometimes shared by threat actors in order to exfiltrate bulk data from an EC2 fleet. If the permissions were modified, verify the snapshot was not shared with an unauthorized or unexpected AWS account

AWS EC2 Traffic Mirroring Configuration Request

Identifies a successful request to create a traffic mirror filter, rule, session or target. Inspect the configuration and destination to determine whether packet mirroring is active and authorized; creating a filter alone does not prove that traffic was captured.

AWS EC2 VM Export Request

Identifies a CreateInstanceExportTask request, including rejected requests. Check actionResult and the vendor error fields to determine whether AWS accepted it. This event does not prove that an export completed or that data was exfiltrated.

AWS ECS/EKS Container Abuse Detection

Detects suspicious container operations in AWS ECS and EKS including registration of task definitions with privileged mode, host network access, or sensitive volume mounts. Attackers may deploy malicious containers to execute code, escalate privileges, or access host resources.

AWS Glue Development Endpoint Privilege Escalation

Detects creation or modification of AWS Glue development endpoints which can be used for privilege escalation. An attacker can pass an IAM role to a Glue dev endpoint and then use it to execute code with that role's permissions, potentially escalating privileges.

AWS GuardDuty Detector Deletion

Identifies the deletion of an Amazon GuardDuty detector. Upon deletion, GuardDuty stops monitoring the environment and all existing findings are lost

AWS GuardDuty Detector Updated or Disabled

Identifies an UpdateDetector call that disables a GuardDuty detector, stopping threat detection while avoiding a clean DeleteDetector signal.

AWS GuardDuty Finding Filter Added

Identifies PutFilters adding a GuardDuty filter that suppresses specific finding types, blinding the account to particular compromise indicators.

AWS GuardDuty High or Critical Finding

Detects an explicitly identified GuardDuty EventBridge finding with severity from 7 through 10. The CloudWatch log group must receive those EventBridge events; GuardDuty management API calls are not findings. Review the original finding detail for affected resources and threat context.

AWS IAM Assume Role Policy Update

Identifies attempts to modify an AWS IAM Assume Role Policy. An adversary may attempt to modify the AssumeRolePolicy of a misconfigured role in order to gain the privileges of that role

AWS IAM Backdoor Creation Attempts

Detects potential IAM backdoor creation attempts through suspicious IAM user creation, access key generation, or policy attachment activities that could provide persistent access.

AWS IAM Deactivation of MFA Device

Identifies the deactivation of a specified multi-factor authentication (MFA) device and removes it from association with the user name for which it was originally enabled. In AWS Identity and Access Management (IAM), a device must be deactivated before it can be deleted

AWS IAM Login Profile Modified for Another User

Detects UpdateLoginProfile API calls where the modifier is potentially different from the target user. This can indicate an attacker changing another user's password to maintain access or enable console login for a compromised programmatic-only account.

AWS IAM Password Recovery Requested

Identifies AWS IAM password recovery requests. An adversary may attempt to gain unauthorized AWS access by abusing password recovery mechanisms.

Potential false positives

Verify whether the user identity, user agent, and/or hostname should be requesting changes in your environment. Password reset attempts from unfamiliar users should be investigated. If known behavior is causing false positives, it can be exempted from the rule.

AWS IAM Privilege Escalation Path Detection

Detects IAM actions commonly chained for privilege escalation including PassRole, CreatePolicyVersion, AttachUserPolicy, and AssumeRole. Attackers exploit these API calls to elevate their permissions within an AWS environment by creating new policy versions with elevated privileges or assuming roles with higher access.

AWS KMS Key Disabled or Deletion Scheduled

Identifies DisableKey or ScheduleKeyDeletion, rendering a customer master key unusable so protected data becomes unreadable, a destructive or ransomware hallmark.

AWS KMS Key Material Import or Deletion

Identifies successful ImportKeyMaterial or DeleteImportedKeyMaterial requests. Verify the key, authorization and dependent resources. These are key-management operations and do not by themselves prove encryption of data or ransomware. Review key lifecycle and recovery plans before taking any containment action.

AWS Lambda Function URL Backdoor Creation

Detects creation of Lambda function URLs which provide direct HTTPS endpoints to Lambda functions. Attackers may create function URL configurations as backdoor endpoints for persistent access, command execution, or data exfiltration without requiring API Gateway.

AWS Management Console Brute Force of Root User Identity

Identifies a high number of failed authentication attempts to the AWS management console for the Root user identity. An adversary may attempt to brute force the password for the Root user identity, as it has complete access to all services and resources for the AWS account

AWS Management Console Root Login

Identifies a successful login to the AWS Management Console by the Root user.

Adversaries may obtain and abuse credentials of a cloud account as a means of gaining Initial Access, Persistence, Privilege Escalation, or Defense Evasion.

Compromised credentials for cloud accounts can be used to harvest sensitive data from online storage accounts and databases.

Potential false positives

It’s strongly recommended that the root user is not used for everyday tasks, including the administrative ones. Verify whether the IP address, location, and/or hostname should be logging in as root in your environment. Unfamiliar root logins should be investigated immediately. If known behavior is causing false positives, it can be exempted from the rule.

AWS Mass Resource Deletion

Detects mass deletion of AWS resources which could indicate destructive attack or insider threat. Monitors for multiple delete operations across various AWS services within a 10-minute window.

AWS RDS Cluster Deletion

Identifies the deletion of an Amazon Relational Database Service (RDS) Aurora database cluster or global database cluster

AWS RDS Deletion Protection Disabled or Master Password Changed

Identifies ModifyDBInstance that disables DeletionProtection or changes MasterUserPassword, preparing destructive deletion or seizing database control.

AWS RDS Publicly Accessible Restore Requested

Identifies a successful database restore request explicitly setting publiclyAccessible to true. Verify the task status, subnet routing, security groups and approval. Request acceptance does not prove that the database is reachable from the internet or that any data was exfiltrated.

AWS RDS Security Group Internet Ingress Added

Detects a successful AuthorizeDBSecurityGroupIngress request allowing 0.0.0.0/0. Revoking a permission does not add exposure. Review the database security group and business justification.

AWS RDS Snapshot Export Request Accepted

Identifies a successful StartExportTask API request. Review the export destination and task status; API acceptance does not establish that the export completed.

AWS RDS Snapshot Sharing for Data Exfiltration

Detects RDS snapshot sharing modifications that could be used for data exfiltration. Attackers may share database snapshots with external AWS accounts to exfiltrate sensitive data, or make snapshots public to download them from an attacker-controlled account.

AWS Root Account Usage Without MFA

Detects usage of AWS root account without Multi-Factor Authentication (MFA). Root account usage should be avoided for daily operations and must always use MFA when necessary. This is a critical security violation as the root account has unrestricted access to all AWS resources.

AWS Root Login Without MFA

Identifies attempts to login to AWS as the root user without using multi-factor authentication (MFA). Amazon AWS best practices indicate that the root user should be protected by MFA

AWS Route 53 DNS Hijacking Attempt

Detects potential DNS hijacking attempts through unauthorized changes to Route 53 DNS records. This rule monitors for ChangeResourceRecordSets operations that could indicate an attacker modifying DNS entries to redirect traffic.

AWS Route 53 Domain Transfer Lock Disabled

Identifies when a transfer lock was removed from a Route 53 domain. It is recommended to refrain from performing this action unless intending to transfer the domain to a different registrar

AWS Route 53 Domain Transfer Request Accepted

Identifies an accepted Route 53 Domains request to transfer a domain to another AWS account. Check the destination account and transfer status; request acceptance does not establish that the transfer completed.

AWS S3 Bucket Configuration Deletion

Identifies the deletion of various Amazon Simple Storage Service (S3) bucket configuration components

AWS S3 Bucket Server Access Logging Disabled

Identifies PutBucketLogging with no target bucket, disabling server-access logging on a bucket and removing the per-object audit trail.

AWS S3 Bucket Versioning Suspended

Detects when S3 bucket versioning is suspended via PutBucketVersioning. Disabling versioning is a cloud ransomware preparation technique that prevents recovery of overwritten or deleted objects from version history.

AWS S3 Bulk Data Exfiltration Detected

Detects bulk GetObject operations from S3 buckets indicating potential data exfiltration. When an attacker gains access to AWS credentials, they may attempt to download large amounts of data from S3 buckets in a short period.

AWS S3 Public Access Controls Relaxed

Detects a successful public ACL change or removal/relaxation of bucket Public Access Block controls. Account-level and bucket policies can still prevent public access. Review the effective controls before concluding that data is exposed.

AWS SAML Role Assumption After Provider Change

Detects a successful AssumeRoleWithSAML request after a successful SAML provider creation or update in the same AWS account and collector within 24 hours. The provider change can be by a different administrator. This sequence requires investigation and does not prove that a SAML assertion was forged.

AWS SSM SendCommand Remote Execution

Detects AWS Systems Manager SendCommand API calls which allow remote command execution on EC2 instances. Attackers with sufficient IAM permissions abuse SSM to execute commands across multiple instances without SSH/RDP access.

AWS SSO Identity Provider Configuration Changed

Detects changes to AWS SSO identity provider configuration including directory association and external IdP configuration changes. Modifying the identity provider can enable impersonation of any user in the organization.

AWS STS Token Abuse Detection

Detects potential abuse of AWS STS AssumeRole operations. This rule identifies when roles are assumed from unusual IP addresses or when there are multiple role assumptions in a short time period, which could indicate lateral movement or privilege escalation. The rule specifically flags AssumeRole operations performed without MFA authentication.

AWS Secrets Manager Suspicious Access Pattern

Detects unusual access patterns to AWS Secrets Manager that could indicate credential theft or unauthorized access attempts. This rule monitors for multiple GetSecretValue or BatchGetSecretValue operations from the same source within a short time window, which may indicate an attacker attempting to harvest credentials.

AWS Security Group Modifications

Detects modifications to AWS security groups that could weaken network security posture. Monitors for changes that add permissive rules or remove restrictive rules, particularly those allowing unrestricted access (0.0.0.0/0 or ::/0).

AWS Security Hub Disabled

Identifies DisableSecurityHub, turning off the central threat dashboard and aggregate findings, an anti-forensics step.

AWS SecurityHub Finding Evasion

Detects attempts to suppress or manipulate AWS SecurityHub findings through BatchUpdateFindings, DeleteInsight, or UpdateFindings operations. Attackers use these to hide evidence of their activities from security monitoring.

AWS Systems Manager Command Request Accepted

Identifies a successful SendCommand API request. Review the document, parameters and target instances for authorization, and inspect command execution results separately. API acceptance does not establish command execution success.

AWS Systems Manager Session Abuse

Detects suspicious use of AWS Systems Manager (SSM) for remote access including StartSession, SendCommand, and SendSSHPublicKey. Attackers use SSM to establish interactive sessions or execute commands on EC2 instances without requiring direct SSH access or security group changes.

AWS TruffleHog Credential Scanning Detected

Detects TruffleHog credential scanning tool activity in AWS CloudTrail logs identified by its characteristic user agent string. TruffleHog is used to scan for exposed secrets and credentials in AWS environments.

AWS Unusual API Call Patterns

Detects unusual API call patterns in AWS that may indicate unauthorized access or reconnaissance activities. This rule triggers when multiple sensitive API calls are made from the same source IP within a short time window, suggesting potential enumeration or discovery activities by attackers.

AWS WAF Access Control List Deletion

Identifies the deletion of a specified AWS Web Application Firewall (WAF) access control list

AWS WAF Rule or Rule Group Deletion

Identifies the deletion of a specified AWS Web Application Firewall (WAF) rule or rule group

AWS WAF and Shield Rule Modifications

Detects modifications or deletions of AWS WAF web ACLs and rules which could be used to remove web application protections before launching an attack. Attackers may disable WAF rules to allow malicious traffic or remove rate limiting protections.

Repeated AWS Administrative Policy Attachments

Detects repeated successful attachment of the AWS-managed AdministratorAccess or IAMFullAccess policy to a role or user by the same actor and account within one hour. Review whether these grants were approved. The request alone does not identify a Lambda execution role.

Repeated AWS SSO Permission Set Changes

Detects repeated successful permission-set creation, policy attachment, deletion or inline-policy updates in AWS SSO from the same source and account. These administrative operations do not establish a failed login or compromised SSO session.

Repeated AWS VPC Flow Log Deletion Requests

Detects repeated successful DeleteFlowLogs API requests from the same source and AWS account within 24 hours. Review the flow-log identifiers and any per-resource failures in the response. API success alone does not prove that every requested log was deleted.

Repeated ECS Task API Requests by an Assumed Role

Detects repeated successful DescribeTaskDefinition, RunTask or StartTask requests by an assumed role. These CloudTrail management calls do not establish access to the task credential endpoint or credential theft. Review the role, task and requested operations for authorization.

Repeated Malformed AWS IAM Trust Policy Updates

Detects repeated UpdateAssumeRolePolicy requests rejected with MalformedPolicyDocument or MalformedPolicyDocumentException by the same actor and account. This is a rejected trust-policy update, not a failed AssumeRole authentication or password brute force. Review the proposed policy and administrative workflow.