Cloud & Containers

Cloud Account Takeover

Stolen credentials, tokens, or recovery paths control a cloud identity.

IntermediateIaaSSaaS
PRIMARY MAPPINGT1078.004

Initial Access / Privilege Escalation

01
CONCEPT

What is it?

Cloud Account Takeover is an attack pattern in which stolen credentials, tokens, or recovery paths control a cloud identity. The learning goal is to identify the trust boundary being abused, the attacker’s sequence, and the telemetry that records it.

02
MENTAL MODEL

How it works in plain English

Use this three-part model to understand the attack without memorizing a tool name.

01 / TRUST BEING ABUSED

The control plane trusts an identity, API token, workload role, policy, image, or orchestrator request.

02 / ATTACKER ACTION

The attacker uses legitimate APIs or workload access to read secrets, alter permissions, create resources, or move data.

03 / DEFENDER VIEW

Cloud investigation is API-centric: identify the principal, credential type, source, method name, target resource, authorization decision, and follow-on calls.

Worked example: how the sequence may look

01 / Starting condition

Access to or influence over the relevant IaaS, SaaS environment.

02 / Attacker action

Trigger the condition that enables cloud account takeover.

03 / Observable evidence

Cloud control-plane audit logs, identity events, API calls, and configuration history.

04 / Detection decision

Baseline the affected identity, host, application, service, or protocol.

05 / Possible outcome

If successful, the attacker may continue with Cloud IAM Privilege Escalation. This is a possibility, not a guaranteed next step.

This is a defensive learning scenario. Real incidents vary, and the listed outcome is only one possible path.

03
ATTACKER OBJECTIVE

Why do attackers use it?

  • Advance an objective associated with initial access / privilege escalation.

  • Exploit a weak control, unsafe default, exposed service, trusted relationship, or human decision.

  • Create access, gain privilege, evade defenses, steal information, or disrupt operations.

04
REQUIRED CONDITIONS

Prerequisites

  • Access to or influence over the relevant IaaS, SaaS environment.

  • Knowledge of the target’s technology, identities, configuration, or user behavior.

  • A weakness, misconfiguration, exposed interface, stolen secret, or social opportunity.

  • A path to observe success and continue toward the objective.

05
ATTACK CHAIN

Attack path possibilities

Attack chains are not fixed. These links show common possibilities to investigate before and after this behavior.

06
SEQUENCE

Step-by-step attack flow

01

Reconnoiter the target and identify the exposed trust boundary.

02

Prepare the infrastructure, lure, input, credential, or payload required.

03

Trigger the condition that enables cloud account takeover.

04

Confirm access or effect while attempting to avoid controls.

05

Use the result for follow-on access, execution, collection, movement, persistence, or impact.

06

Remove evidence, rotate infrastructure, or repeat against other targets.

07
FRAMEWORK

MITRE ATT&CK mapping

TECHNIQUET1078.004
NAMECloud Account Takeover
TACTICInitial Access / Privilege Escalation
VERIFY ON MITRE ↗

ATT&CK is updated over time. Verify the live technique before operationalizing a detection.

08
EVIDENCE

Logs and artifacts

  • Cloud control-plane audit logs, identity events, API calls, and configuration history.

  • Container runtime, orchestrator audit, workload, registry, and network-flow telemetry.

  • Unusual role assumption, secret access, public exposure, or data-transfer activity.

09
RAW TELEMETRY

Realistic log examples

REPRESENTATIVE, SANITIZED EXAMPLES

These records use realistic field names and formats but synthetic organizations, users, addresses, and identifiers. A single event is not proof; correlate time, identity, source, target, and the resulting action.

LOG SOURCEAWS CloudTrail

Who called which AWS API, from where, with what session and MFA context.

{"eventTime":"2026-07-16T14:31:22Z","eventSource":"sts.amazonaws.com",
"eventName":"AssumeRole","awsRegion":"ap-south-1",
"sourceIPAddress":"203.0.113.48","userAgent":"aws-cli/2.17",
"userIdentity":{"type":"IAMUser","arn":"arn:aws:iam::111122223333:user/build"},
"requestParameters":{"roleArn":"arn:aws:iam::111122223333:role/Admin"},
"responseElements":{"credentials":{"accessKeyId":"ASIA...REDACTED"}}}
HOW TO INTERPRET IT FOR CLOUD ACCOUNT TAKEOVER

Baseline the affected identity, host, application, service, or protocol.

LOG SOURCEAzure Activity / Entra audit

Azure control-plane operation, caller, resource, correlation, and authorization result.

2026-07-16T14:32:16Z OperationName="Create or Update Role Assignment"
Category=Administrative ActivityStatus=Succeeded
Caller=build@corp.example CallerIpAddress=203.0.113.48
ResourceId=/subscriptions/0000.../providers/Microsoft.Authorization/roleAssignments/7c21...
CorrelationId=1a2c8432-1f10-4b38-a962-14577bb41d0e
HOW TO INTERPRET IT FOR CLOUD ACCOUNT TAKEOVER

Detect rare combinations of source, target, action, timing, volume, and result.

LOG SOURCEGoogle Cloud Audit Log

Principal, API method, resource, caller IP, and authorization context.

{"timestamp":"2026-07-16T14:33:02Z",
"logName":"projects/acme-prod/logs/cloudaudit.googleapis.com%2Factivity",
"protoPayload":{"authenticationInfo":{"principalEmail":"build@acme-prod.iam.gserviceaccount.com"},
"methodName":"SetIamPolicy","resourceName":"projects/acme-prod",
"requestMetadata":{"callerIp":"203.0.113.48","callerSuppliedUserAgent":"gcloud/480.0"}}}
HOW TO INTERPRET IT FOR CLOUD ACCOUNT TAKEOVER

Correlate identity, endpoint, network, application, and control-plane evidence across the sequence.

LOG SOURCEKubernetes API audit

API verb, principal, source, resource, namespace, decision, and response code.

{"apiVersion":"audit.k8s.io/v1","stage":"ResponseComplete",
"requestURI":"/api/v1/namespaces/prod/secrets/db-admin",
"verb":"get","user":{"username":"system:serviceaccount:ci:runner"},
"sourceIPs":["10.24.18.77"],"objectRef":{"resource":"secrets",
"namespace":"prod","name":"db-admin"},"responseStatus":{"code":200}}
HOW TO INTERPRET IT FOR CLOUD ACCOUNT TAKEOVER

Prioritize activity followed by successful access, privilege change, execution, or data movement.

READ EACH LOG WITH FIVE QUESTIONS
  1. Who or what identity acted?
  2. From which device, process, IP, or workload?
  3. What target and operation were involved?
  4. Did it fail, succeed, or change state?
  5. What correlated event happened immediately before and after?
10
ANALYTICS

Detection logic / SIEM queries

  • Baseline the affected identity, host, application, service, or protocol.

  • Detect rare combinations of source, target, action, timing, volume, and result.

  • Correlate identity, endpoint, network, application, and control-plane evidence across the sequence.

  • Prioritize activity followed by successful access, privilege change, execution, or data movement.

Pseudo-SIEMBehavioral detection template
FROM security_events
WHERE event_time > now() - 30m
  AND behavior = "Cloud Account Takeover"
GROUP BY source, target, user
HAVING count(*) > baseline(source, target, user)
   OR rare(action, source, target) = true
CORRELATE WITH success, privilege_change, execution, data_access
11
HARDENING

Mitigations

  • Reduce exposed attack surface and remove unused services, identities, features, and trust relationships.

  • Apply least privilege, strong authentication, secure defaults, segmentation, and timely patching.

  • Validate untrusted input and independently authorize sensitive operations.

  • Centralize and protect relevant logs; test detections with controlled simulations.

  • Maintain a response playbook for containment, credential rotation, evidence preservation, and recovery.

12
IN THE WILD

Real-world examples

  • Cloud Account Takeover appears in opportunistic campaigns and targeted intrusions when the required condition exists.

  • Real incidents usually combine it with credential theft, phishing, exploitation, persistence, or exfiltration.

  • The best case studies describe the full attack chain rather than an isolated tool or indicator.

13
REVIEW

Common questions

01What makes Cloud Account Takeover possible?

A trust assumption fails: an identity, input, component, network message, user decision, or software relationship is accepted without enough verification.

02What should a defender collect first?

Start with logs closest to the decision point, then add identity, endpoint, network, and control-plane context.

03How should I study this attack?

Learn the concept, identify prerequisites, map observable steps, write a detection hypothesis, and validate it safely in a lab.

Primary verification sources