MITRE ATT&CK technique · T1098.001
Account Manipulation: Additional Cloud Credentials
How do attackers add backup credentials to AWS accounts? After compromising an identity with IAM write access, adversaries add credentials to other users — new access keys, console passwords, SSH keys or certificates — so that losing the original credential does not end their access.
For leadership
Executive takeaway
How it works in AWS
- The adversary lists users and their existing keys (each IAM user may have up to two keys).
- They call iam:CreateAccessKey or iam:CreateLoginProfile for a user they do not own — often a dormant service user.
- The new credential is used later, from different infrastructure.
Sub-technique of T1098 Account Manipulation.
What the attacker needs
- iam:CreateAccessKey / CreateLoginProfile / UpdateLoginProfile on other users
Relevant AWS API calls
| API / event | Why it matters |
|---|---|
| iam:CreateAccessKey | Creates a new long-term key for a specified user. |
| iam:CreateLoginProfile / UpdateLoginProfile | Adds or resets a console password. |
| iam:UploadSSHPublicKey | Adds a key for CodeCommit access. |
A risky permission is not an observed attack
Non-admin identities that hold iam:CreateAccessKey on Resource "*"; IAM users with no owner.
CreateAccessKey where requestParameters.userName differs from the caller; new key used shortly afterwards from a new network.
Preventive and limiting controls
- Restrict IAM write permissions
- SCP denying key creation except for break-glass
- Prefer roles so few IAM users exist
Mapped cloud controls
- preventiveLeast-privilege identity policies
Only identities with iam:CreateAccessKey (or equivalent) on other users can add credentials to them.
- restrictiveService control policy guardrails
An SCP can deny creating IAM user access keys outside an approved break-glass path.
- detectiveAlerting on identity and logging changes
CreateAccessKey and CreateLoginProfile are management events and are high-signal when unexpected.
- recoveryCredential compromise response runbook
Persistence via extra credentials is only removed if responders enumerate and revoke every key and session.
See assumptions and validation evidence in the Cloud Control Explorer →
Detection signals and telemetry
- CreateAccessKey / CreateLoginProfile by a different principal than the target user — CloudTrail management events
MITRE detection strategies (official, cloud-relevant analytics)
- DET0531 Detection Strategy for Additional Cloud Credentials in IaaS/IdP/SaaS
- AN1469 (Identity Provider): Addition of credentials (keys, app passwords, x.509 certs) to existing cloud accounts, service principals, or OAuth apps via portal or API by non-standard identities or IP ranges.Log sources: azure:audit — Add service principal credentials, app password added, app role assignment
- AN1470 (IaaS): Cloud API usage to create/import SSH keys or generate new access keys (CreateAccessKey, ImportKeyPair, CreateLoginProfile) from non-console access or unusual principals.Log sources: AWS:CloudTrail — CreateAccessKey, ImportKeyPair, CreateLoginProfile, CreateKeyPair; gcp:audit — iam.serviceAccounts.keys.create, os-login.sshPublicKeys.add
- AN1471 (SaaS): Credential-related configuration changes in productivity apps, such as API key creation in Google Workspace, app tokens in Slack, or user-level OAuth credentials in M365.Log sources: gcp:audit — API Key Created, OAuth Client Registered; m365:unified — Set-Mailbox, Set-AppPassword, Add-MailboxPermission
MITRE mitigations
Official MITRE description (excerpt)
Adversaries may add adversary-controlled credentials to a cloud account to maintain persistent access to victim accounts and instances within the environment.
For example, adversaries may add credentials for Service Principals and Applications in addition to existing legitimate credentials in Azure / Entra ID. These credentials include both x509 keys and passwords. With sufficient permissions, there are a variety of ways to add credentials including the Azure Portal, Azure command line interface, and Azure or Az PowerShell modules.
Common misconceptions
- “Users can only have one access key.”
- IAM users can have two active keys; adversaries exploit the spare slot.
Attack scenarios using this technique
Compromised AWS access key
Cloud credential persistence
AI agent with excessive AWS permissions
References
MITRE ATT&CK® and ATT&CK® are registered trademarks of The MITRE Corporation. ATT&CK data © 2026 The MITRE Corporation. This work is reproduced and distributed with the permission of The MITRE Corporation (ATT&CK v20.1). Securitain is not affiliated with or endorsed by MITRE. Explanations, scenarios and control mappings on this page are Securitain’s own analysis. Sources and licensing