Overview of role recommendations

Role recommendations help you identify and remove excess permissions from your principals, improving your resources' security configurations.

Overview of role recommendations

Role recommendations are generated by the IAM recommender. The IAM recommender is one of the recommenders that Recommender offers.

Each role recommendation suggests that you remove or replace a role that gives your principals excess permissions. At scale, these recommendations help you enforce the principle of least privilege by ensuring that principals have only the permissions that they actually need.

The IAM recommender identifies excess permissions using policy insights. Policy insights are ML-based findings about a principal's permission usage.

Some recommendations are also associated with lateral movement insights. These insights identify roles that allow service accounts in one project to impersonate service accounts in another project. For more information, see How lateral movement insights are generated.

How policy insights are generated

Policy insights highlight which permissions in a given role binding a principal has actually used.

The IAM recommender generates policy insights by comparing a principal's total number of permissions with the permissions that the principal used in the last 90 days. If the role binding was created fewer than 90 days ago, the IAM recommender looks at the principal's permission usage in the time since the binding was created.

There are a few ways in which a principal can use a permission:

  • Directly, by calling an API that requires the permission

    For example, the roles.list method in the IAM REST API requires the iam.roles.list permission. When you call the roles.list method, you use the iam.roles.list permission.

    Similarly, when you call the testIamPermissions method for a resource, you effectively use all of the permissions that you are testing.

  • Indirectly, by using the Google Cloud console to work with Google Cloud resources

    For example, in the Google Cloud console, you can edit a Compute Engine virtual machine (VM) instance, which requires different permissions based on which settings you change. However, the Google Cloud console also displays the existing settings, which requires the compute.instances.get permission.

    As a result, when you edit a VM instance in the Google Cloud console, you use the compute.instances.get permission.

The IAM recommender also uses machine learning to identify permissions in a principal's current role that the principal is likely to need in the future, even if the principal did not use those permissions recently. For more information, see Machine learning for policy insights on this page.

Policy insights are not generated for all IAM role bindings. For more information about why a role binding might not have a policy insight, see Availability on this page.

To learn how to manage policy insights, see Manage policy insights for projects, folders, and organizations or Manage policy insights for Cloud Storage buckets.

Machine learning for policy insights

In some cases, a principal is likely to need certain permissions that are included in their current roles, but that they haven't used recently. To identify these permissions, the IAM recommender uses a machine learning (ML) model when generating policy insights.

This machine learning model is trained on multiple sets of signals:

  • Common co-occurrence patterns in the observed history: The fact that a user used permission A, B, and C in the past provides a hint that A, B, and C might be related in some way and that they are needed together to carry out a task on Google Cloud. If the ML model observes this pattern frequently enough, the next time a different user uses permission A and B, the model will suggest that the user might need permission C as well.

  • Domain knowledge as encoded in the role definitions: IAM provides hundreds of different predefined roles that are service-specific. If a predefined role contains a set of permissions, it is a strong signal that those permissions should be granted together.

In addition to these signals, the model also uses word embedding to calculate how semantically similar the permissions are. Semantically similar permissions will be "close" to each other after embedding, and more likely to be granted together. For example, bigquery.datasets.get and bigquery.tables.list will be very close to each other after embedding.

All data used in the IAM recommender machine learning pipeline has k-anonymity, meaning that individuals in the anonymized data set cannot be re-identified. To achieve this level of anonymity, we drop all personally identifiable information (PII) such as the user ID related to each permission usage pattern. Then we drop all usage patterns that do not show up frequently enough across Google Cloud. The global model is trained on this anonymized data.

The global model can be further customized for each organization using federated learning, a machine learning process that trains machine learning models without exporting data.

How role recommendations are generated

If a policy insight indicates that a principal does not need all of the permissions in their role, the IAM recommender assesses the role to determine if it could be revoked, or if there is another role that's a better fit. If the role can be revoked, the IAM recommender generates a role recommendation to revoke the role. If there is another role that's a better fit, the IAM recommender generates a role recommendation to replace the role with a suggested role. This suggested role could be a new custom role, an existing custom role, or one or more predefined roles. Except in the case of recommendations for Google-managed service accounts, a role recommendation never suggests a change that increases a principal's level of access.

Role recommendations are generated based on only IAM access controls. They do not take into account other kinds of access controls, like access control lists (ACLs) and Kubernetes role-based access control (RBAC). If you use other types of access controls, take extra care when you review your recommendations, and consider how those access controls relate to your allow policies.

Additionally, role recommendations are not generated for all IAM role bindings. For more information about why a role binding might not have a role recommendation, see Availability on this page.

Observation period

A role recommendation's observation period is the number of days of permission usage data that the recommendation is based on.

The maximum observation period for role recommendations is 90 days. This means that the IAM recommender uses, at most, the most recent 90 days of permission usage data to generate role recommendations.

The IAM recommender also doesn't start generating role recommendations until it has a certain number of days of permission usage data. This duration is called the minimum observation period. By default, the minimum observation period is 90 days, but, for project-level role recommendations, you can manually set it to 30 days or 60 days. For details, see Configure role recommendation generation. If you set the minimum observation period to fewer than 90 days, you'll get recommendations sooner, but the accuracy of the recommendations might be affected.

If a role binding is older than the minimum observation period but newer than 90 days, the observation period is the lifetime of the role binding.

New custom roles in role recommendations

When the IAM recommender suggests replacements for a role, it always suggests an existing custom role, or one or more predefined roles, that appear to be a better fit for the principal's needs.

If the IAM recommender identifies a common permission usage pattern in your organization that does not map to an existing predefined or custom role, it might also recommend that you create a new project-level custom role. This custom role includes only the recommended permissions. You can modify the custom role recommendation by adding or removing permissions.

If you want to enforce the principle of least privilege as strictly as possible, choose the new custom role. The IAM recommender creates the custom role at the project level. You are responsible for maintaining and updating the custom roles for your projects.

If you prefer to use a Google-managed role, choose the predefined role. Google Cloud updates these roles regularly by adding or removing permissions. To be notified about these updates, subscribe to the news feed for the permissions change log. When you choose the predefined role, the principal continues to have at least a few permissions, and potentially a large number of permissions, that they have not used.

The IAM recommender does not recommend new custom roles in the following cases:

  • The recommendation is for a folder-level or organization-level role.
  • Your organization already has 100 or more custom roles.
  • Your project already has 25 or more custom roles.

Also, the IAM recommender recommends no more than 5 new custom roles per day in each project, and no more than 15 new custom roles across the entire organization.

How lateral movement insights are generated

Lateral movement is when a service account in one project has permission to impersonate a service account in another project. For example, a service account might have been created in project A, but have permissions to impersonate a service account in project B.

These permissions can result in a chain of impersonations across projects that gives principals unintended access to resources. For example, if a principal impersonates the service account in project A, they can use that service account to impersonate the service account in project B. If the service account in project B has permission to impersonate other service accounts in other projects in your organization, the principal can continue to use service account impersonation to move from project to project, gaining permissions as they go.

The IAM recommender generates lateral movement insights by identifying role bindings that fit the following criteria:

  • The principal in the role binding is a service account that was not created in the project.
  • The role includes one of the following permissions, which allow a principal to impersonate a service account:

    • iam.serviceAccounts.actAs
    • iam.serviceAccounts.getAccessToken
    • iam.serviceAccounts.getOpenIdToken
    • iam.serviceAccounts.implicitDelegation
    • iam.serviceAccounts.signBlob
    • iam.serviceAccounts.signJwt

If a role binding fits these criteria, the IAM recommender generates a lateral movement insight for the binding. This insight contains information about the service account's impersonation abilities, including which service accounts it can impersonate and whether it used any impersonation permissions in the last 90 days.

The IAM recommender does not use lateral movement insights on their own to generate new role recommendations. This is because, if a service account is using its impersonation permissions, the IAM recommender can't safely suggest removing them. However, if a role recommendation suggests removing these permissions because they aren't being used, the IAM recommender will link the lateral movement insight to that recommendation. This linkage helps you prioritize role recommendations for service accounts that have powerful, unused impersonation permissions across projects.

To learn how to manage lateral movement insights, see Manage lateral movement insights.

Availability

Policy insights, lateral movement insights, and role recommendations are not generated for every role binding. Read the following sections to understand the role bindings that policy insights, lateral movement insights, and recommendations are generated for.

Policy insight availability

For the IAM recommender to generate a policy insight for a role binding, the following must be true:

  • The IAM allow policy that contains the role binding must be attached to one of the following resources:

    • Cloud Storage bucket
    • BigQuery dataset
    • Project
    • Folder
    • Organization

    The IAM recommender only generates policy insights for roles that are granted on these resources.

  • The role binding must not have a condition. The IAM recommender does not generate policy insights for conditional role bindings.

It can take up to 10 days for the IAM recommender to generate policy insights for a new role binding.

Existing policy insights are refreshed daily based on your recent permission usage. However, the data that the policy insights are based on typically lags up to two days behind your actual usage.

Lateral movement insight availability

Lateral movement insights are generated for role bindings on the following resources:

  • Organizations
  • Folders
  • Projects
  • Service accounts

Role recommendation availability

For the IAM recommender to generate a role recommendation for a role binding, the following must be true:

  • The role binding must have a policy insight associated with it. This policy insight serves as the basis for the recommendation.
  • The role binding must be older than the minimum observation period. This ensures that the IAM recommender has enough usage data to make a recommendation. By default, the minimum observation period is 90 days, but you can manually set it to 30 days or 60 days. For details, see Configure role recommendation generation.
  • If the principal in the role binding is a Google-managed service account, the role binding must be Owner, Editor, or Viewer. The IAM recommender does not generate role recommendations for Google-managed service accounts with other roles. For more details, see Role recommendations for Google-managed service accounts.

If a role binding is too new or does not have any insights, the Analyzed permissions column in the Google Cloud console shows a icon.

There are some cases where the IAM recommender does not generate role recommendations for a role binding, even though it is old enough and has an insight associated with it. This can happen for the following reasons:

  • There are no predefined IAM roles that are more appropriate than the current role. If a principal already has a predefined role that minimizes their permissions, or that includes fewer permissions than other predefined roles, then the IAM recommender cannot recommend a different predefined role.

    You might be able to reduce the principal's permissions by creating a custom role for the principal.

  • The principal is a Google-managed service account, and the role is not a basic role. The IAM recommender only generates role recommendations for Google-managed service accounts if the service account has a basic role (Owner, Editor, or Viewer). For more details, see Role recommendations for Google-managed service accounts.

  • No other principal has the Owner basic role for the project. At least one principal must have the Owner role (roles/owner) for each project. If only one principal has this role, the IAM recommender will not recommend that you revoke or replace the role.

In these cases, the Analyzed permissions column in the Google Cloud console shows the principal's permission usage, but does not have a Recommendation available icon.

Existing role recommendations are refreshed daily based on your recent permission usage. However, the data that the role recommendations are based on typically lags up to two days behind your actual usage.

Priority and severity

Recommendation priority and insight severity help you understand the urgency of a recommendation or insight and prioritize accordingly.

Role recommendation priority

Recommendations are assigned priority levels based their perceived urgency. Priority levels range from P0 (highest priority) to P4 (lowest priority).

A role recommendation's priority depends on the role binding that the recommendation is for:

Recommendation details Priority Explanation
Recommendations for role bindings that grant public access to Cloud Storage buckets P1 Publicly accessible buckets can be accessed by anyone on the internet. Removing public access gives you more control over your data.
Recommendations for role bindings that grant public access to BigQuery datasets P1 Publicly accessible datasets can be accessed by anyone on the internet. Removing public access gives you more control over your data.
Recommendations for role bindings that grant basic roles (Owner, Editor, and Viewer) on a project, folder, or organization P2 Basic roles are highly permissive, and applying recommendations for these roles can greatly reduce your over-granted permissions.
Recommendations that do not grant public access or basic roles P4 Although these recommendations help you reduce excess permissions, they don't remove public access or highly permissive basic roles, which makes them a lower priority.

Insight severity

Insights are assigned severity levels based their perceived urgency. Severity levels can be LOW, MEDIUM, HIGH, or CRITICAL.

A policy insight's severity depends on the role binding that the insight is for:

Insight details Severity Explanation
Insights for role bindings that grant public access to Cloud Storage buckets CRITICAL Publicly accessible buckets can be accessed by anyone on the internet. Removing public access gives you more control over your data.
Insights for role bindings that grant public access to BigQuery datasets CRITICAL Publicly accessible datasets can be accessed by anyone on the internet. Removing public access gives you more control over your data.
Insights for role bindings that grant basic roles (Owner, Editor, and Viewer) on a project, folder, or organization HIGH Basic roles are highly permissive, and addressing insights for these roles can greatly reduce your over-granted permissions.
Insights that do not grant public access or basic roles LOW Although these insights highlight excess permissions, they don't involve public access or highly permissive basic roles, which makes them a lower priority.

All lateral movement insights have a severity of LOW.

How role recommendations are applied

The IAM recommender does not apply recommendations automatically. Instead, you must review your recommendations and decide whether to apply or dismiss them. To learn how to review, apply, and dismiss role recommendations, see one of the following guides:

Audit logging

When you apply or dismiss a recommendation, the IAM recommender creates a log entry. You can view these entries in your recommendations history, or you can view them in your Google Cloud audit logs.

Role recommendation subtypes

Role recommendations are split into several different subtypes based on the action they recommend. If you use the gcloud CLI or the REST API, you can use these subtypes to filter your recommendations.

Subtype Description
REMOVE_ROLE A recommendation to remove the principal's project-level, folder-level, or organization-level role.
REMOVE_ROLE_BIGQUERY_DATASET A recommendation to remove the principal's dataset-level role.
REMOVE_ROLE_STORAGE_BUCKET A recommendation to remove the principal's bucket-level role.
REPLACE_ROLE A recommendation to replace the principal's project-level, folder-level, or organization-level role with a less permissive role. The recommended replacement could be an existing custom role or one or more predefined roles.
REPLACE_ROLE_CUSTOMIZABLE A recommendation to replace the principal's role with a new custom role that's less permissive than their current role.
REPLACE_ROLE_BIGQUERY_DATASET A recommendation to replace the principal's dataset-level role with a less permissive role. The recommended replacement could be an existing custom role or one or more predefined roles.
REPLACE_ROLE_STORAGE_BUCKET A recommendation to replace the principal's bucket-level role with a less permissive role. The recommended replacement could be an existing custom role or one or more predefined roles.
SERVICE_AGENT_WITH_DEFAULT_ROLE A recommendation to replace a Google-managed service account's Owner, Editor, or Viewer role with the role that was automatically granted to the service account when it was created. For more information, see Recommendations for Google-managed service accounts.
SERVICE_AGENT_WITHOUT_DEFAULT_ROLE A recommendation to replace a Google-managed service account's Owner, Editor, or Viewer role with a less permissive role. For more information, see Recommendations for Google-managed service accounts.

Role recommendations for Google-managed service accounts

For Google-managed service accounts, the IAM recommender only provides recommendations for role bindings with basic roles (Owner, Editor, or Viewer).

Recommendations for Google-managed service accounts are divided into two recommendation subtypes.

SERVICE_AGENT_WITH_DEFAULT_ROLE

On creation, some Google-managed service accounts are automatically granted a service agent role to ensure that your Google Cloud services work properly. If you replace this role with a basic role (Owner, Editor, or Viewer), a role recommendation might suggest that you restore the original service agent role to remove excess permissions, even if the service agent role has permissions that are not in the basic role. These recommendations have the subtype SERVICE_AGENT_WITH_DEFAULT_ROLE. They help you safely remove excess permissions while ensuring that all Google Cloud services work properly.

SERVICE_AGENT_WITH_DEFAULT_ROLE recommendations are the only type of recommendation that might suggest roles with permissions not in the current role.

SERVICE_AGENT_WITHOUT_DEFAULT_ROLE

If a Google-managed service account is not automatically granted a role on creation, recommendations for the service account are based exclusively on the permissions that the service account uses. These recommendations have the subtype SERVICE_AGENT_WITHOUT_DEFAULT_ROLE.

Role recommendations in Security Command Center

If you have the Premium tier of Security Command Center, you can view some subtypes of role recommendations as findings in Security Command Center. Each subtype is associated with a detector:

Recommendation subtype Finding category
REMOVE_ROLE Unused IAM role
REPLACE_ROLE IAM role has excessive permissions
SERVICE_AGENT_WITH_DEFAULT_ROLE Service agent role replaced with basic role
SERVICE_AGENT_WITHOUT_DEFAULT_ROLE Service agent granted basic role

For more information about viewing role recommendations in Security Command Center, see IAM recommender in the Security Command Center documentation.

Examples of role recommendations

The following examples show the types of recommendations that you can receive.

Revoke an existing role

The user my-user@example.com was granted the Browser role on a project. The Browser role includes six permissions that allow the user to view resources in the project. However, during the past 90 days, my-user@example.com hasn't viewed any resources.

Therefore, the IAM recommender generates a role recommendation suggesting that you revoke the Browser role from my-user@example.com:

Console

gcloud

{
  "associatedInsights": [
    {
      "insight": "projects/123456789012/locations/global/insightTypes/google.iam.policy.Insight/insights/86c14538-dcfd-4326-afe5-ee8ac921e06a"
    }
  ],
  "content": {
    "operationGroups": [
      {
        "operations": [
          {
            "action": "remove",
            "path": "/iamPolicy/bindings/*/members/*",
            "pathFilters": {
              "/iamPolicy/bindings/*/condition/expression": "",
              "/iamPolicy/bindings/*/members/*": "user:my-user@example.com",
              "/iamPolicy/bindings/*/role": "roles/browser"
            },
            "resource": "//cloudresourcemanager.googleapis.com/projects/123456789012",
            "resourceType": "cloudresourcemanager.googleapis.com/Project"
          }
        ]
      }
    ],
    "overview": {
      "member": "user:my-user@example.com",
      "removedRole": "roles/browser",
      "resource": "//cloudresourcemanager.googleapis.com/projects/123456789012"
    }
  },
  "description": "This role has not been used during the observation window.",
  "etag": "\"9fc3241da8bfab51\"",
  "lastRefreshTime": "2022-05-20T07:00:00Z",
  "name": "projects/123456789012/locations/global/recommenders/google.iam.policy.Recommender/recommendations/fe512038-5455-49g1-8f9c-752e31c8c154",
  "primaryImpact": {
    "category": "SECURITY",
    "securityProjection": {
      "details": {
        "revokedIamPermissionsCount": 6
      }
    }
  },
  "priority": "P4",
  "recommenderSubtype": "REMOVE_ROLE",
  "stateInfo": {
    "state": "ACTIVE"
  }
}

REST

{
  "name": "projects/123456789012/locations/global/recommenders/google.iam.policy.Recommender/recommendations/fe512038-5455-49g1-8f9c-752e31c8c154",
  "description": "This role has not been used during the observation window.",
  "lastRefreshTime": "2022-05-20T07:00:00Z",
  "primaryImpact": {
    "category": "SECURITY",
    "securityProjection": {
      "details": {
        "revokedIamPermissionsCount": 6
      }
    }
  },
  "content": {
    "operationGroups": [
      {
        "operations": [
          {
            "action": "remove",
            "resourceType": "cloudresourcemanager.googleapis.com/Project",
            "resource": "//cloudresourcemanager.googleapis.com/projects/123456789012",
            "path": "/iamPolicy/bindings/*/members/*",
            "pathFilters": {
              "/iamPolicy/bindings/*/condition/expression": "",
              "/iamPolicy/bindings/*/members/*": "user:my-user@example.com",
              "/iamPolicy/bindings/*/role": "roles/browser"
            }
          }
        ]
      }
    ],
    "overview": {
      "resource": "//cloudresourcemanager.googleapis.com/projects/123456789012",
      "member": "user:my-user@example.com",
      "removedRole": "roles/browser"
    }
  },
  "stateInfo": {
    "state": "ACTIVE"
  },
  "etag": "\"9fc3241da8bfab51\"",
  "recommenderSubtype": "REMOVE_ROLE",
  "associatedInsights": [
    {
      "insight": "projects/123456789012/locations/global/insightTypes/google.iam.policy.Insight/insights/86c14538-dcfd-4326-afe5-ee8ac921e06a"
    }
  ],
  "priority": "P4"
}

Replace an existing role

A service account was granted the Editor role (roles/editor) on a project. This basic role includes more than 3,000 permissions and grants extensive access to the project. However, during the past 90 days, the service account has only used a few of those permissions.

Therefore, the IAM recommender generates a role recommendation suggesting that you revoke the Editor role and replace it with a combination of two other roles, which removes thousands of excess permissions:

Console

gcloud

{
  "associatedInsights": [
    {
      "insight": "projects/123456789012/locations/global/insightTypes/google.iam.policy.Insight/insights/3d4ef3d6-bdf0-4330-975d-c65cb929c44d"
    }
  ],
  "content": {
    "operationGroups": [
      {
        "operations": [
          {
            "action": "add",
            "path": "/iamPolicy/bindings/*/members/-",
            "pathFilters": {
              "/iamPolicy/bindings/*/condition/expression": "",
              "/iamPolicy/bindings/*/role": "roles/iam.serviceAccountUser"
            },
            "resource": "//cloudresourcemanager.googleapis.com/projects/123456789012",
            "resourceType": "cloudresourcemanager.googleapis.com/Project",
            "value": "user:my-user@example.com"
          },
          {
            "action": "add",
            "path": "/iamPolicy/bindings/*/members/-",
            "pathFilters": {
              "/iamPolicy/bindings/*/condition/expression": "",
              "/iamPolicy/bindings/*/role": "roles/storage.objectAdmin"
            },
            "resource": "//cloudresourcemanager.googleapis.com/projects/123456789012",
            "resourceType": "cloudresourcemanager.googleapis.com/Project",
            "value": "user:my-user@example.com"
          },
          {
            "action": "remove",
            "path": "/iamPolicy/bindings/*/members/*",
            "pathFilters": {
              "/iamPolicy/bindings/*/condition/expression": "",
              "/iamPolicy/bindings/*/members/*": "user:my-user@example.com",
              "/iamPolicy/bindings/*/role": "roles/editor"
            },
            "resource": "//cloudresourcemanager.googleapis.com/projects/123456789012",
            "resourceType": "cloudresourcemanager.googleapis.com/Project"
          }
        ]
      }
    ],
    "overview": {
      "addedRoles": [
        "roles/iam.serviceAccountUser",
        "roles/storage.objectAdmin"
      ],
      "member": "user:my-user@example.com",
      "minimumObservationPeriodInDays": "0",
      "removedRole": "roles/editor",
      "resource": "//cloudresourcemanager.googleapis.com/projects/123456789012"
    }
  },
  "description": "Replace the current role with smaller predefined roles to cover the permissions needed.",
  "etag": "\"0da9a354c2a83d96\"",
  "lastRefreshTime": "2022-06-22T07:00:00Z",
  "name": "projects/123456789012/locations/global/recommenders/google.iam.policy.Recommender/recommendations/4637db3d-dba5-45eb-95ac-b4ee4b4cd14e",
  "primaryImpact": {
    "category": "SECURITY",
    "securityProjection": {
      "details": {
        "revokedIamPermissionsCount": 2998
      }
    }
  },
  "priority": "P2",
  "recommenderSubtype": "REPLACE_ROLE",
  "stateInfo": {
    "state": "ACTIVE"
  }
}

REST

{
  "name": "projects/123456789012/locations/global/recommenders/google.iam.policy.Recommender/recommendations/4637db3d-dba5-45eb-95ac-b4ee4b4cd14e",
  "description": "Replace the current role with smaller predefined roles to cover the permissions needed.",
  "lastRefreshTime": "2022-06-22T07:00:00Z",
  "primaryImpact": {
    "category": "SECURITY",
    "securityProjection": {
      "details": {
        "revokedIamPermissionsCount": 2998
      }
    }
  },
  "content": {
    "operationGroups": [
      {
        "operations": [
          {
            "action": "add",
            "resourceType": "cloudresourcemanager.googleapis.com/Project",
            "resource": "//cloudresourcemanager.googleapis.com/projects/123456789012",
            "path": "/iamPolicy/bindings/*/members/-",
            "value": "user:my-user@example.com",
            "pathFilters": {
              "/iamPolicy/bindings/*/condition/expression": "",
              "/iamPolicy/bindings/*/role": "roles/iam.serviceAccountOwner"
            }
          },
          {
            "action": "add",
            "resourceType": "cloudresourcemanager.googleapis.com/Project",
            "resource": "//cloudresourcemanager.googleapis.com/projects/123456789012",
            "path": "/iamPolicy/bindings/*/members/-",
            "value": "user:my-user@example.com",
            "pathFilters": {
              "/iamPolicy/bindings/*/condition/expression": "",
              "/iamPolicy/bindings/*/role": "roles/storage.objectAdmin"
            }
          },
          {
            "action": "remove",
            "resourceType": "cloudresourcemanager.googleapis.com/Project",
            "resource": "//cloudresourcemanager.googleapis.com/projects/123456789012",
            "path": "/iamPolicy/bindings/*/members/*",
            "pathFilters": {
              "/iamPolicy/bindings/*/condition/expression": "",
              "/iamPolicy/bindings/*/members/*": "user:my-user@example.com",
              "/iamPolicy/bindings/*/role": "roles/editor"
            }
          }
        ]
      }
    ],
    "overview": {
      "resource": "//cloudresourcemanager.googleapis.com/projects/123456789012",
      "member": "user:my-user@example.com",
      "removedRole": "roles/editor",
      "addedRoles": [
        "roles/iam.serviceAccountUser",
        "roles/storage.objectAdmin"
      ],
      "minimumObservationPeriodInDays": "0"
    }
  },
  "stateInfo": {
    "state": "ACTIVE"
  },
  "etag": "\"0da9a354c2a83d96\"",
  "recommenderSubtype": "REPLACE_ROLE",
  "associatedInsights": [
    {
      "insight": "projects/123456789012/locations/global/insightTypes/google.iam.policy.Insight/insights/3d4ef3d6-bdf0-4330-975d-c65cb929c44d"
    }
  ],
  "priority": "P2"
}

Create a custom role

The user my-user@example.com was granted the Cloud Trace Admin role (roles/cloudtrace.admin) on a project. The role includes more than 10 permissions, but a policy insight indicates that, during the past 90 days, my-user@example.com used only 4 of those permissions.

Therefore, the IAM recommender generates a role recommendation suggesting that you create a custom role that includes only the permissions that my-user@example.com actually used:

Console

gcloud

The REPLACE_ROLE_CUSTOMIZABLE subtype indicates that the IAM recommender recommends creating a custom role with the used permissions. To see the used permissions, get the associated policy insight.

{
  "associatedInsights": [
    {
      "insight": "projects/123456789012/locations/global/insightTypes/google.iam.policy.Insight/insights/2799dc04-b12e-4cf6-86aa-d81907d31f58"
    }
  ],
  "associatedResourceNames": [
    "//cloudresourcemanager.googleapis.com/projects/123456789012"
  ],
  "content": {
    "operationGroups": [
      {
        "operations": [
          {
            "action": "add",
            "path": "/iamPolicy/bindings/*/members/-",
            "pathFilters": {
              "/iamPolicy/bindings/*/condition/expression": "",
              "/iamPolicy/bindings/*/role": "roles/cloudtrace.user"
            },
            "resource": "//cloudresourcemanager.googleapis.com/projects/123456789012",
            "resourceType": "cloudresourcemanager.googleapis.com/Project",
            "value": "user:my-user@example.com"
          },
          {
            "action": "remove",
            "path": "/iamPolicy/bindings/*/members/*",
            "pathFilters": {
              "/iamPolicy/bindings/*/condition/expression": "",
              "/iamPolicy/bindings/*/members/*": "user:my-user@example.com",
              "/iamPolicy/bindings/*/role": "roles/cloudtrace.admin"
            },
            "resource": "//cloudresourcemanager.googleapis.com/projects/123456789012",
            "resourceType": "cloudresourcemanager.googleapis.com/Project"
          }
        ]
      }
    ],
    "overview": {
      "minimumObservationPeriodInDays": "0"
    }
  },
  "description": "Replace the current role with a smaller role to cover the permissions needed.",
  "etag": "\"c7f57a4725d32d66\"",
  "lastRefreshTime": "2022-06-22T07:00:00Z",
  "name": "projects/123456789012/locations/global/recommenders/google.iam.policy.Recommender/recommendations/ba1fc977-fddd-3856-a829-f69649ae8075",
  "originalContent": {},
  "primaryImpact": {
    "category": "SECURITY",
    "securityProjection": {
      "details": {},
      "revokedIamPermissionsCount": 1
    }
  },
  "priority": "P4",
  "recommenderSubtype": "REPLACE_ROLE_CUSTOMIZABLE",
  "stateInfo": {
    "state": "ACTIVE"
  },
  "targetResources": [
    "//cloudresourcemanager.googleapis.com/projects/123456789012"
  ]
}

REST

The REPLACE_ROLE_CUSTOMIZABLE subtype indicates that the IAM recommender recommends creating a custom role with the used permissions. To see the used permissions, get the associated policy insight.

{
  "name": "projects/123456789012/locations/global/recommenders/google.iam.policy.Recommender/recommendations/ba1fc977-fddd-3856-a829-f69649ae8075",
  "description": "Replace the current role with a smaller role to cover the permissions needed.",
  "lastRefreshTime": "2022-06-22T07:00:00Z",
  "primaryImpact": {
    "category": "SECURITY",
    "securityProjection": {
      "details": {
        "revokedIamPermissionsCount": 1
      }
    }
  },
  "content": {
    "operationGroups": [
      {
        "operations": [
          {
            "action": "add",
            "resourceType": "cloudresourcemanager.googleapis.com/Project",
            "resource": "//cloudresourcemanager.googleapis.com/projects/123456789012",
            "path": "/iamPolicy/bindings/*/members/-",
            "value": "user:my-user@example.com",
            "pathFilters": {
              "/iamPolicy/bindings/*/condition/expression": "",
              "/iamPolicy/bindings/*/role": "roles/cloudtrace.user"
            }
          },
          {
            "action": "remove",
            "resourceType": "cloudresourcemanager.googleapis.com/Project",
            "resource": "//cloudresourcemanager.googleapis.com/projects/123456789012",
            "path": "/iamPolicy/bindings/*/members/*",
            "pathFilters": {
              "/iamPolicy/bindings/*/condition/expression": "",
              "/iamPolicy/bindings/*/members/*": "user:my-user@example.com",
              "/iamPolicy/bindings/*/role": "roles/cloudtrace.admin"
            }
          }
        ]
      }
    ],
    "overview": {
      "minimumObservationPeriodInDays": "0"
    }
  },
  "stateInfo": {
    "state": "ACTIVE"
  },
  "etag": "\"c7f57a4725d32d66\"",
  "recommenderSubtype": "REPLACE_ROLE_CUSTOMIZABLE",
  "associatedInsights": [
    {
      "insight": "projects/123456789012/locations/global/insightTypes/google.iam.policy.Insight/insights/2799dc04-b12e-4cf6-86aa-d81907d31f58"
    }
  ],
  "priority": "P4"
}

The role recommendation also suggests another option, which is to replace the existing role with the Cloud Trace User role (roles/cloudtrace.user). This predefined role includes slightly fewer permissions than the Cloud Trace Admin role.

Role replacement with permissions suggested by machine learning

A service account was granted the Editor role (roles/editor) on a project. This basic role includes more than 3,000 permissions and grants extensive access to a project. However, a policy insight indicates that, during the past 90 days, the service account has used fewer than 10 permissions.

The policy insight also highlights several permissions that service account is likely to need in the future. The IAM recommender identified these permissions using machine learning.

The IAM recommender generates a role recommendation suggesting that you revoke the Editor role and replace it with the Storage Object Admin role (roles/storage.objectAdmin), which grants full control of objects in a Cloud Storage bucket. This change removes thousands of excess permissions, while still including both the permissions the service account used and the permissions that the service account is likely to need in the future:

Console

The IAM recommender uses a Machine learning icon to identify permissions that were added based on the IAM recommender's machine learning rather than on permission usage. In this example, the resourcemanager.projects.get permission was recommended based on machine learning:

gcloud

Permissions that were added based on the IAM recommender's machine learning rather than on permission usage aren't listed in the recommendation itself. Instead, they're listed in the policy insights associated with the recommendation. All ML-based permissions are listed in the inferredPermissions field of the insight. In this example, the resourcemanager.projects.get permission was recommended based on machine learning:

associatedRecommendations:
- recommendation: projects/123456789012/locations/global/recommenders/google.iam.policy.Recommender/recommendations/0573b702-96a5-4622-a916-c762e7b0731f
category: SECURITY
content:
  condition:
    description: ''
    expression: ''
    location: ''
    title: ''
  currentTotalPermissionsCount: '5069'
  exercisedPermissions:
  - permission: storage.objects.create
  - permission: storage.objects.delete
  - permission: storage.objects.get
  - permission: storage.objects.list
  inferredPermissions:
  - permission: resourcemanager.projects.get
  member: serviceAccount:my-service-account@my-project.iam.gserviceaccount.com
  role: roles/editor
description: 4 of the permissions in this role binding were used in the past 90 days.
etag: '"d3cdec23cc712bd0"'
insightSubtype: PERMISSIONS_USAGE
lastRefreshTime: '2020-07-11T07:00:00Z'
name: projects/123456789012/locations/global/insightTypes/google.iam.policy.Insight/insights/0d3ce433-f067-4e78-b6ae-03d7d1f6f040
observationPeriod: 7776000s
stateInfo:
  state: ACTIVE
targetResources:
- //cloudresourcemanager.googleapis.com/projects/123456789012
severity: HIGH

To learn how to get a policy insight, see Review policy insights.

REST

Permissions that were added based on the IAM recommender's machine learning rather than on permission usage aren't listed in the recommendation itself. Instead, they're listed in the policy insights associated with the recommendation. All ML-based permissions are listed in the inferredPermissions field of the insight. In this example, the resourcemanager.projects.get permission was recommended based on machine learning:

{
  "name": "projects/123456789012/locations/global/insightTypes/google.iam.policy.Insight/insights/07841f74-02ce-4de8-bbe6-fc4eabb68568",
  "description": "4 of the permissions in this role binding were used in the past 90 days.",
  "content": {
    "role": "roles/editor",
    "member": "serviceAccount:my-service-account@my-project.iam.gserviceaccount.com",
    "condition": {
      "expression": "",
      "title": "",
      "description": "",
      "location": ""
    },
    "exercisedPermissions": [
      {
        "permission": "storage.objects.create"
      },
      {
        "permission": "storage.objects.delete"
      },
      {
        "permission": "storage.objects.get"
      },
      {
        "permission": "storage.objects.list"
      }
    ],
    "inferredPermissions": [
      {
        "permission": "resourcemanager.projects.get"
      }
    ],
    "currentTotalPermissionsCount": "5069"
  },
  "lastRefreshTime": "2020-07-12T07:00:00Z",
  "observationPeriod": "7776000s",
  "stateInfo": {
    "state": "ACTIVE"
  },
  "category": "SECURITY",
  "associatedRecommendations": [
    {
      "recommendation": "projects/123456789012/locations/global/recommenders/google.iam.policy.Recommender/recommendations/b1932220-867d-43d1-bd74-fb95876ab656"
    }
  ],
  "targetResources": [
    "//cloudresourcemanager.googleapis.com/projects/123456789012"
  ],
  "insightSubtype": "PERMISSIONS_USAGE",
  "etag": "\"d3cdec23cc712bd0\"",
  "severity": "HIGH"
}

To learn how to get a policy insight, see Review policy insights.

What's next