July Recap: New AWS Sensitive Permissions and Services

4 mins to read

As AWS continues to evolve, new services and permissions are frequently introduced to enhance functionality and security. This blog provides a comprehensive recap of new sensitive permissions and services added in July 2024. Our intention in sharing this is to flag the most important releases to keep your eye on and update your permissions and access control policies accordingly.

Existing Services with New Sensitive Permissions

Amazon Bedrock

Service Type: Artificial Intelligence & Machine Learning

Permission: bedrock:UpdatePrompt
  • Action: Grants permission to update a prompt
  • Mitre Tactic: Impact
  • Why it’s sensitive: Allows changes to prompts affecting model behavior.  Unauthorized updates can lead to incorrect outputs, security risks, data leaks, and negative user interactions.
Permission: bedrock:UpdateFlow
  • Action: Grants permission to update a prompt flow
  • Mitre Tactic: Impact
  • Why it’s sensitive: Allows modifications to the AI workflows including changing a prompt in a flow or even adding nodes to a flow after the initial CreateFlow. Some node types can allow the AI model to access other resources in your cloud, including retrieving data from S3 and invoking Lambda functions. 

Amazon DataZone

Service Type: Data Management

Permission: datazone:DeleteAssetFilter
  • Action: Grants permission to delete an asset filter
  • Mitre Tactic: Defense Invasion
  • Why it’s sensitive: Asset filters restrict access to specific elements of the source data the Datazone portal is peering through, impacting what gets shown. Altering or removing these protections can expose intentionally restricted sensitive data.
Permission: datazone:UpdateAssetFilter
  • Action: Grants permission to update an asset filter
  • Mitre Tactic: Defense Evasion
  • Why it’s sensitive: Asset filters restrict access to specific elements of the source data the Datazone portal is peering through, impacting what gets shown. Altering or removing these protections can expose intentionally restricted sensitive data.

AWS CloudHSM

Service Type: Security and Compliance

Permission: cloudhsm:PutResourcePolicy
  • Action: Grants permission to attach a policy to an AWS CloudHSM resource
  • Mitre Tactic: Persistence
  • Why it’s sensitive:  Allows users to set or modify resource policies for AWS CloudHSM. Incorrect or malicious policy changes could grant unauthorized access to cryptographic resources, potentially compromising sensitive data and keys.

AWS CodeConnections

Service Type: Development and DevOps Tools

Permission: codeconnections:CreateSyncConfiguration
  • Action: Grants permission to create a template sync config
  • Mitre Tactic: Persistence
  • Why it’s sensitive:  Allows to sync with external repositories. Modifications could lead to data corruption, loss, or inconsistency, impacting the integrity of the data being synchronized.

AWS CodeStar Connections

Service Type: Development and DevOps Tools

Permission: codestar-connections:CreateHost
  • Action: Grants permission to create a host resource
  • Mitre Tactic: Persistence
  • Why it’s sensitive:  Allows connection to external systems, which could be untrusted or malicious.

AWS Service Catalog

Service Type: Infrastructure Management

Permission: servicecatalog:AssociatePrincipalWithPortfolio
  • Action: Grants permission to associate an IAM principal with a portfolio, giving the specified principal access to any products associated with the specified portfolio
  • Mitre Tactic: Privilege Escalation
  • Why it’s sensitive:  Associating IAM principals to a portfolio grants them access to the shared portfolio.
Permission: servicecatalog:UpdateProvisioningArtifact
  • Action: Grants permission to update the metadata fields of an existing provisioning artifact
  • Mitre Tactic: Defense Evasion
  • Why it’s sensitive:  Updates the active version of the specific product, which includes reverting to a previous version.  
Permission: servicecatalog:DeleteConstraint
  • Action: Grants permission to remove and delete an existing constraint from an associated product and portfolio
  • Mitre Tactic: Persistence
  • Why it’s sensitive: Allows removal of security control and governance policies.
Permission: servicecatalog:DisableAWSOrganizationsAccess
  • Action: Grants permission to disable portfolio sharing through AWS Organizations feature
  • Mitre Tactic: Privilege Escalation
  • Why it’s sensitive:  Allows existing accounts to retain access, only new accounts are prevented access.  
Permission: servicecatalog:UpdateProvisionedProduct
  • Action: Grants permission to update an existing provisioned product
  • Mitre Tactic: Execution
  • Why it’s sensitive:  Allows updates to live, provisioned AWS resources which could cause interruptions.
Permission: servicecatalog:ExecuteProvisionedProductPlan
  • Action: Grants permission to execute a provisioned product plan
  • Mitre Tactic: Execution
  • Why it’s sensitive:  Allows creation or update of provisioned products, which could cause service disruptions or downtime.
Permission: servicecatalog:UpdateProvisionedProductProperties
  • Action: Grants permission to update the properties of an existing provisioned product
  • Mitre Tactic: Privilege Escalation
  • Why it’s sensitive: Allows updates to live instances of AWS resources which could lead to service disruptions.
Permission: servicecatalog:UpdateConstraint
  • Action: Grants permission to update the metadata fields of an existing constraint
  • Mitre Tactic: Persistence
  • Why it’s sensitive:  Allows altering or removing parameters of the constraint, including the role.
Permission: servicecatalog:ExecuteProvisionedProductServiceAction
  • Action: Grants permission to execute a provisioned product plan
  • Mitre Tactic: Execution
  • Why it’s sensitive:  Allows users without full access to AWS services to perform self-service actions.
Permission: servicecatalog:ProvisionProduct
  • Action: Grants permission to provision a product with a specified provisioning artifact and launch parameters
  • Mitre Tactic: Execution
  • Why it’s sensitive:  Allows provisioning of products which can lead to unplanned cost or unauthorized access.

New Services

AWS App Studio

Service Type: Natural Language and Speech Services

Permission: appstudio:StartEnablementJob
  • Action: Grants permission to submit an enablement job
  • Mitre Tactic: Persistence
  • Why it’s sensitive: Creates an IAM policy for accessing AWS services, uses or creates a CodeCatalyst space for projects at no extra cost, and deploys a CloudFormation stack with IAM roles for DynamoDB resources. 

Conclusion

If you’re an AWS user, your cloud is always changing. This means a constantly evolving attack surface for you to secure. As new permissions are released for pre-existing services, by default, your users gain access to that permission. If it is a sensitive permission, this can be risky.  Access to sensitive permissions should be restricted to only those human and machine identities that need them.

To reduce the risk resulting from new services, your teams should update any SCPs and IAM policies used to restrict access to services your teams aren’t using.

If you’re interested in managing sensitive permissions and securing AWS services efficiently, look into our Cloud Permissions Firewall.

secure sensitive permissions