# Bedrock IAM actions (bedrock) and their access levels

Facts read 16 Nov 2026. Page reviewed 16 Nov 2026.

Bedrock has 269 IAM actions under the prefix bedrock in AWS's service authorization reference fetched on 16 Nov 2026: 48 List, 83 Read, 135 Write, 1 Permissions management and 2 Tagging. A wildcard bedrock:* in an Allow statement grants every one of them, and the tables below list each action with its resource types.

## Permissions management: 1 actions

| Action | Resource types | Condition keys |
| --- | --- | --- |
| `bedrock:AllowVendedLogDeliveryForResource` | knowledge-base | 0 |

Permissions management actions of Bedrock, fetched 16 Nov 2026.

## Write: 135 actions

| Action | Resource types | Condition keys |
| --- | --- | --- |
| `bedrock:AssociateAgentCollaborator` | agent | 0 |
| `bedrock:AssociateAgentKnowledgeBase` | agent, knowledge-base | 0 |
| `bedrock:AssociateThirdPartyKnowledgeBase` | * | 1 |
| `bedrock:BatchDeleteAdvancedPromptOptimizationJob` | advanced-prompt-optimization-job | 0 |
| `bedrock:BatchDeleteEvaluationJob` | evaluation-job | 0 |
| `bedrock:CancelAutomatedReasoningPolicyBuildWorkflow` | automated-reasoning-policy | 0 |
| `bedrock:CancelInvoke` | project | 0 |
| `bedrock:CopyBlueprintStage` | blueprint | 0 |
| `bedrock:CreateAdvancedPromptOptimizationJob` | application-inference-profile, foundation-model, inference-profile | 2 |
| `bedrock:CreateAgent` | * | 2 |
| `bedrock:CreateAgentActionGroup` | agent | 2 |
| `bedrock:CreateAgentAlias` | agent | 2 |
| `bedrock:CreateAutomatedReasoningPolicy` | * | 2 |
| `bedrock:CreateAutomatedReasoningPolicyTestCase` | automated-reasoning-policy | 0 |
| `bedrock:CreateAutomatedReasoningPolicyVersion` | automated-reasoning-policy | 2 |
| `bedrock:CreateBlueprint` | * | 2 |
| `bedrock:CreateBlueprintVersion` | blueprint | 0 |
| `bedrock:CreateCustomModel` | * | 2 |
| `bedrock:CreateCustomModelDeployment` | custom-model | 2 |
| `bedrock:CreateDataAutomationLibrary` | * | 2 |
| `bedrock:CreateDataAutomationProject` | blueprint, data-automation-library | 2 |
| `bedrock:CreateDataSource` | knowledge-base | 0 |
| `bedrock:CreateDataSourceToken` | * | 0 |
| `bedrock:CreateEvaluationJob` | custom-model, default-prompt-router, foundation-model, prompt-router | 2 |
| `bedrock:CreateFlow` | * | 2 |
| `bedrock:CreateFlowAlias` | flow | 2 |
| `bedrock:CreateFlowVersion` | flow | 0 |
| `bedrock:CreateFoundationModelAgreement` | * | 0 |
| `bedrock:CreateGuardrail` | automated-reasoning-policy, automated-reasoning-policy-version, guardrail-profile | 2 |
| `bedrock:CreateGuardrailVersion` | guardrail | 0 |
| `bedrock:CreateInferenceProfile` | application-inference-profile, foundation-model, inference-profile | 2 |
| `bedrock:CreateInvocation` | session | 0 |
| `bedrock:CreateKnowledgeBase` | * | 2 |
| `bedrock:CreateMarketplaceModelEndpoint` | * | 0 |
| `bedrock:CreateModelCopyJob` | custom-model | 2 |
| `bedrock:CreateModelCustomizationJob` | custom-model, foundation-model | 2 |
| `bedrock:CreateModelEvaluationJob` | custom-model, foundation-model | 2 |
| `bedrock:CreateModelImportJob` | * | 2 |
| `bedrock:CreateModelInvocationJob` | custom-model, foundation-model, model-invocation-job | 2 |
| `bedrock:CreatePrompt` | * | 2 |
| `bedrock:CreatePromptRouter` | application-inference-profile, foundation-model, inference-profile | 2 |
| `bedrock:CreatePromptVersion` | prompt | 2 |
| `bedrock:CreateProvisionedModelThroughput` | custom-model, foundation-model | 2 |
| `bedrock:CreateRetrievalInstruction` | * | 2 |
| `bedrock:CreateRetrievalInstructionVersion` | retrieval-instruction | 0 |
| `bedrock:CreateSession` | * | 2 |
| `bedrock:DeleteAgent` | agent | 0 |
| `bedrock:DeleteAgentActionGroup` | agent | 0 |
| `bedrock:DeleteAgentAlias` | agent-alias | 0 |
| `bedrock:DeleteAgentMemory` | agent-alias | 0 |
| `bedrock:DeleteAgentVersion` | agent | 0 |
| `bedrock:DeleteAutomatedReasoningPolicy` | automated-reasoning-policy, automated-reasoning-policy-version | 0 |
| `bedrock:DeleteAutomatedReasoningPolicyBuildWorkflow` | automated-reasoning-policy | 0 |
| `bedrock:DeleteAutomatedReasoningPolicyTestCase` | automated-reasoning-policy | 0 |
| `bedrock:DeleteBlueprint` | blueprint | 0 |
| `bedrock:DeleteCustomModel` | custom-model | 0 |
| `bedrock:DeleteCustomModelDeployment` | custom-model-deployment | 0 |
| `bedrock:DeleteDataAutomationLibrary` | data-automation-library | 0 |
| `bedrock:DeleteDataAutomationProject` | data-automation-project | 0 |
| `bedrock:DeleteDataSource` | knowledge-base | 0 |
| `bedrock:DeleteEnforcedGuardrailConfiguration` | * | 0 |
| `bedrock:DeleteFlow` | flow | 0 |
| `bedrock:DeleteFlowAlias` | flow-alias | 0 |
| `bedrock:DeleteFlowVersion` | flow | 0 |
| `bedrock:DeleteFoundationModelAgreement` | * | 0 |
| `bedrock:DeleteGuardrail` | guardrail | 0 |
| `bedrock:DeleteImportedModel` | imported-model | 0 |
| `bedrock:DeleteInferenceProfile` | application-inference-profile | 0 |
| `bedrock:DeleteInvoke` | project | 0 |
| `bedrock:DeleteKnowledgeBase` | knowledge-base | 0 |
| `bedrock:DeleteKnowledgeBaseDocuments` | knowledge-base | 0 |
| `bedrock:DeleteMarketplaceModelAgreement` | * | 0 |
| `bedrock:DeleteMarketplaceModelEndpoint` | bedrock-marketplace-model-endpoint | 0 |
| `bedrock:DeleteModelInvocationLoggingConfiguration` | * | 0 |
| `bedrock:DeletePrompt` | prompt, prompt-version | 0 |
| `bedrock:DeletePromptRouter` | prompt-router | 0 |
| `bedrock:DeleteProvisionedModelThroughput` | provisioned-model | 0 |
| `bedrock:DeleteResourcePolicy` | custom-model, guardrail, guardrail-profile, knowledge-base | 0 |
| `bedrock:DeleteRetrievalInstruction` | retrieval-instruction | 0 |
| `bedrock:DeleteSession` | session | 0 |
| `bedrock:DeregisterMarketplaceModelEndpoint` | bedrock-marketplace-model-endpoint | 0 |
| `bedrock:DisassociateAgentCollaborator` | agent | 0 |
| `bedrock:DisassociateAgentKnowledgeBase` | agent, knowledge-base | 0 |
| `bedrock:EndSession` | session | 0 |
| `bedrock:IngestKnowledgeBaseDocuments` | knowledge-base | 0 |
| `bedrock:InvokeBlueprintOptimizationAsync` | blueprint, data-automation-profile | 2 |
| `bedrock:InvokeBlueprintRecommendationAsync` | data-automation-profile | 0 |
| `bedrock:InvokeBuilder` | * | 0 |
| `bedrock:InvokeDataAutomation` | blueprint, data-automation-profile, data-automation-project | 0 |
| `bedrock:InvokeDataAutomationAsync` | blueprint, data-automation-profile, data-automation-project | 2 |
| `bedrock:InvokeDataAutomationLibraryIngestionJob` | data-automation-library | 2 |
| `bedrock:PrepareAgent` | agent | 0 |
| `bedrock:PrepareFlow` | flow | 0 |
| `bedrock:PutAccountDataRetention` | * | 1 |
| `bedrock:PutEnforcedGuardrailConfiguration` | * | 0 |
| `bedrock:PutFoundationModelEntitlement` | * | 0 |
| `bedrock:PutInvocationStep` | session | 0 |
| `bedrock:PutModelInvocationLoggingConfiguration` | * | 0 |
| `bedrock:PutResourcePolicy` | custom-model, guardrail, guardrail-profile, knowledge-base | 2 |
| `bedrock:PutUseCaseForModelAccess` | * | 0 |
| `bedrock:RegisterMarketplaceModelEndpoint` | bedrock-marketplace-model-endpoint | 0 |
| `bedrock:Rerank` | * | 0 |
| `bedrock:RetrieveAndGenerate` | * | 0 |
| `bedrock:StartAutomatedReasoningPolicyBuildWorkflow` | automated-reasoning-policy | 0 |
| `bedrock:StartAutomatedReasoningPolicyTestWorkflow` | automated-reasoning-policy | 0 |
| `bedrock:StartFlowExecution` | flow, flow-alias | 0 |
| `bedrock:StartIngestionJob` | knowledge-base | 0 |
| `bedrock:StopAdvancedPromptOptimizationJob` | advanced-prompt-optimization-job | 0 |
| `bedrock:StopEvaluationJob` | evaluation-job | 0 |
| `bedrock:StopFlowExecution` | flow, flow-alias, flow-execution | 0 |
| `bedrock:StopIngestionJob` | knowledge-base | 0 |
| `bedrock:StopModelCustomizationJob` | model-customization-job | 0 |
| `bedrock:StopModelInvocationJob` | model-invocation-job | 0 |
| `bedrock:UpdateAgent` | agent | 0 |
| `bedrock:UpdateAgentActionGroup` | agent | 0 |
| `bedrock:UpdateAgentAlias` | agent-alias | 0 |
| `bedrock:UpdateAgentCollaborator` | agent | 0 |
| `bedrock:UpdateAgentKnowledgeBase` | agent, knowledge-base | 0 |
| `bedrock:UpdateAutomatedReasoningPolicy` | automated-reasoning-policy | 0 |
| `bedrock:UpdateAutomatedReasoningPolicyAnnotations` | automated-reasoning-policy | 0 |
| `bedrock:UpdateAutomatedReasoningPolicyTestCase` | automated-reasoning-policy | 0 |
| `bedrock:UpdateBlueprint` | blueprint | 0 |
| `bedrock:UpdateCustomModelDeployment` | custom-model, custom-model-deployment | 0 |
| `bedrock:UpdateDataAutomationLibrary` | data-automation-library | 0 |
| `bedrock:UpdateDataAutomationProject` | blueprint, data-automation-project, data-automation-project | 0 |
| `bedrock:UpdateDataSource` | knowledge-base | 0 |
| `bedrock:UpdateFlow` | flow | 0 |
| `bedrock:UpdateFlowAlias` | flow-alias | 0 |
| `bedrock:UpdateGuardrail` | automated-reasoning-policy, automated-reasoning-policy-version, guardrail, guardrail-profile | 0 |
| `bedrock:UpdateKnowledgeBase` | knowledge-base | 0 |
| `bedrock:UpdateMarketplaceModelEndpoint` | bedrock-marketplace-model-endpoint | 0 |
| `bedrock:UpdatePrompt` | prompt | 0 |
| `bedrock:UpdateProvisionedModelThroughput` | custom-model, foundation-model, provisioned-model | 0 |
| `bedrock:UpdateRetrievalInstruction` | retrieval-instruction | 0 |
| `bedrock:UpdateSession` | session | 0 |

Write actions of Bedrock, fetched 16 Nov 2026.

## Tagging: 2 actions

| Action | Resource types | Condition keys |
| --- | --- | --- |
| `bedrock:TagResource` | advanced-prompt-optimization-job, agent, agent-alias, application-inference-profile, async-invoke, automated-reasoning-policy, automated-reasoning-policy-version, blueprint, blueprint-optimization-invocation, custom-model, custom-model-deployment, data-automation-invocation-job, data-automation-library, data-automation-library-ingestion-job, data-automation-project, evaluation-job, flow, flow-alias, guardrail, imported-model, knowledge-base, model-copy-job, model-customization-job, model-evaluation-job, model-import-job, model-invocation-job, prompt, prompt-router, prompt-version, provisioned-model, retrieval-instruction, session | 2 |
| `bedrock:UntagResource` | advanced-prompt-optimization-job, agent, agent-alias, application-inference-profile, async-invoke, automated-reasoning-policy, automated-reasoning-policy-version, blueprint, blueprint-optimization-invocation, custom-model, custom-model-deployment, data-automation-invocation-job, data-automation-library, data-automation-library-ingestion-job, data-automation-project, evaluation-job, flow, flow-alias, guardrail, imported-model, knowledge-base, model-copy-job, model-customization-job, model-evaluation-job, model-import-job, model-invocation-job, prompt, prompt-router, prompt-version, provisioned-model, retrieval-instruction, session | 1 |

Tagging actions of Bedrock, fetched 16 Nov 2026.

## Read: 83 actions

| Action | Resource types | Condition keys |
| --- | --- | --- |
| `bedrock:AgenticRetrieveStream` | * | 0 |
| `bedrock:ApplyGuardrail` | guardrail, guardrail-profile | 0 |
| `bedrock:CallWithBearerToken` | * | 1 |
| `bedrock:CheckIngestedDocumentAcl` | * | 0 |
| `bedrock:CountTokens` | foundation-model | 0 |
| `bedrock:DetectGeneratedContent` | foundation-model | 0 |
| `bedrock:ExportAutomatedReasoningPolicyVersion` | automated-reasoning-policy, automated-reasoning-policy-version | 0 |
| `bedrock:GenerateQuery` | * | 0 |
| `bedrock:GetAccountDataRetention` | * | 0 |
| `bedrock:GetAdvancedPromptOptimizationJob` | advanced-prompt-optimization-job | 0 |
| `bedrock:GetAgent` | agent | 0 |
| `bedrock:GetAgentActionGroup` | agent | 0 |
| `bedrock:GetAgentAlias` | agent-alias | 0 |
| `bedrock:GetAgentCollaborator` | agent | 0 |
| `bedrock:GetAgentKnowledgeBase` | agent, knowledge-base | 0 |
| `bedrock:GetAgentMemory` | agent-alias | 0 |
| `bedrock:GetAgentVersion` | agent | 0 |
| `bedrock:GetAsyncInvoke` | async-invoke | 0 |
| `bedrock:GetAutomatedReasoningPolicy` | automated-reasoning-policy, automated-reasoning-policy-version | 0 |
| `bedrock:GetAutomatedReasoningPolicyAnnotations` | automated-reasoning-policy | 0 |
| `bedrock:GetAutomatedReasoningPolicyBuildWorkflow` | automated-reasoning-policy | 0 |
| `bedrock:GetAutomatedReasoningPolicyBuildWorkflowResultAssets` | automated-reasoning-policy | 0 |
| `bedrock:GetAutomatedReasoningPolicyNextScenario` | automated-reasoning-policy | 0 |
| `bedrock:GetAutomatedReasoningPolicyTestCase` | automated-reasoning-policy | 0 |
| `bedrock:GetAutomatedReasoningPolicyTestResult` | automated-reasoning-policy | 0 |
| `bedrock:GetBlueprint` | blueprint | 0 |
| `bedrock:GetBlueprintOptimizationStatus` | blueprint-optimization-invocation | 0 |
| `bedrock:GetBlueprintRecommendation` | * | 0 |
| `bedrock:GetCustomModel` | custom-model | 0 |
| `bedrock:GetCustomModelDeployment` | custom-model-deployment | 0 |
| `bedrock:GetDataAutomationLibrary` | data-automation-library | 0 |
| `bedrock:GetDataAutomationLibraryEntity` | data-automation-library | 0 |
| `bedrock:GetDataAutomationLibraryIngestionJob` | data-automation-library, data-automation-library-ingestion-job | 0 |
| `bedrock:GetDataAutomationProject` | data-automation-project | 0 |
| `bedrock:GetDataAutomationStatus` | data-automation-invocation-job | 0 |
| `bedrock:GetDataSource` | knowledge-base | 0 |
| `bedrock:GetDataSourceAuthorizationUrl` | * | 0 |
| `bedrock:GetDocumentContent` | * | 0 |
| `bedrock:GetEvaluationJob` | evaluation-job | 0 |
| `bedrock:GetExecutionFlowSnapshot` | flow, flow-alias, flow-execution | 0 |
| `bedrock:GetFlow` | flow | 0 |
| `bedrock:GetFlowAlias` | flow-alias | 0 |
| `bedrock:GetFlowExecution` | flow, flow-alias, flow-execution | 0 |
| `bedrock:GetFlowVersion` | flow | 0 |
| `bedrock:GetFoundationModel` | foundation-model | 0 |
| `bedrock:GetFoundationModelAvailability` | * | 0 |
| `bedrock:GetGuardrail` | guardrail | 0 |
| `bedrock:GetImportedModel` | imported-model | 0 |
| `bedrock:GetInferenceProfile` | application-inference-profile, inference-profile | 0 |
| `bedrock:GetIngestedDocumentAcl` | * | 0 |
| `bedrock:GetIngestionJob` | knowledge-base | 0 |
| `bedrock:GetInvocationStep` | session | 0 |
| `bedrock:GetInvoke` | project | 0 |
| `bedrock:GetKnowledgeBase` | knowledge-base | 0 |
| `bedrock:GetKnowledgeBaseDocuments` | knowledge-base | 0 |
| `bedrock:GetMarketplaceModelEndpoint` | bedrock-marketplace-model-endpoint | 0 |
| `bedrock:GetModelCopyJob` | model-copy-job | 0 |
| `bedrock:GetModelCustomizationJob` | model-customization-job | 0 |
| `bedrock:GetModelEvaluationJob` | model-evaluation-job | 0 |
| `bedrock:GetModelImportJob` | model-import-job | 0 |
| `bedrock:GetModelInvocationJob` | model-invocation-job | 0 |
| `bedrock:GetModelInvocationLoggingConfiguration` | * | 0 |
| `bedrock:GetPrompt` | prompt, prompt-version | 0 |
| `bedrock:GetPromptRouter` | default-prompt-router, prompt-router | 0 |
| `bedrock:GetProvisionedModelThroughput` | provisioned-model | 0 |
| `bedrock:GetResourcePolicy` | custom-model, guardrail, guardrail-profile, knowledge-base | 0 |
| `bedrock:GetRetrievalInstruction` | retrieval-instruction | 0 |
| `bedrock:GetSession` | session | 0 |
| `bedrock:GetUseCaseForModelAccess` | * | 0 |
| `bedrock:InvokeAgent` | agent-alias | 0 |
| `bedrock:InvokeAutomatedReasoningPolicy` | automated-reasoning-policy, automated-reasoning-policy-version | 0 |
| `bedrock:InvokeFlow` | flow-alias | 0 |
| `bedrock:InvokeGuardrailChecks` | * | 0 |
| `bedrock:InvokeInlineAgent` | * | 1 |
| `bedrock:InvokeModel` | application-inference-profile, async-invoke, bedrock-marketplace-model-endpoint, custom-model-deployment, default-prompt-router, foundation-model, imported-model, inference-profile, project, prompt-router, provisioned-model, system-tool | 8 |
| `bedrock:InvokeModelWithResponseStream` | application-inference-profile, bedrock-marketplace-model-endpoint, custom-model-deployment, default-prompt-router, foundation-model, imported-model, inference-profile, prompt-router, provisioned-model, system-tool | 4 |
| `bedrock:InvokeTool` | system-tool | 0 |
| `bedrock:ListMarketplaceModelEndpoints` | * | 0 |
| `bedrock:ListTagsForResource` | advanced-prompt-optimization-job, agent, agent-alias, application-inference-profile, async-invoke, automated-reasoning-policy, automated-reasoning-policy-version, blueprint, blueprint-optimization-invocation, custom-model, custom-model-deployment, data-automation-invocation-job, data-automation-library, data-automation-library-ingestion-job, data-automation-project, evaluation-job, flow, flow-alias, guardrail, imported-model, knowledge-base, model-copy-job, model-customization-job, model-evaluation-job, model-import-job, model-invocation-job, prompt, prompt-router, prompt-version, provisioned-model, retrieval-instruction, session | 0 |
| `bedrock:OptimizePrompt` | * | 0 |
| `bedrock:RenderPrompt` | prompt, prompt-version | 0 |
| `bedrock:Retrieve` | knowledge-base | 0 |
| `bedrock:ValidateFlowDefinition` | * | 0 |

Read actions of Bedrock, fetched 16 Nov 2026.

## List: 48 actions

| Action | Resource types | Condition keys |
| --- | --- | --- |
| `bedrock:ListAdvancedPromptOptimizationJobs` | * | 0 |
| `bedrock:ListAgentActionGroups` | agent | 0 |
| `bedrock:ListAgentAliases` | agent | 0 |
| `bedrock:ListAgentCollaborators` | agent | 0 |
| `bedrock:ListAgentKnowledgeBases` | agent | 0 |
| `bedrock:ListAgentVersions` | agent | 0 |
| `bedrock:ListAgents` | * | 0 |
| `bedrock:ListAsyncInvokes` | * | 0 |
| `bedrock:ListAutomatedReasoningPolicies` | automated-reasoning-policy | 0 |
| `bedrock:ListAutomatedReasoningPolicyBuildWorkflows` | automated-reasoning-policy | 0 |
| `bedrock:ListAutomatedReasoningPolicyTestCases` | automated-reasoning-policy | 0 |
| `bedrock:ListAutomatedReasoningPolicyTestResults` | automated-reasoning-policy | 0 |
| `bedrock:ListBlueprints` | data-automation-project | 0 |
| `bedrock:ListCustomModelDeployments` | * | 0 |
| `bedrock:ListCustomModels` | * | 0 |
| `bedrock:ListDataAutomationLibraries` | data-automation-project | 0 |
| `bedrock:ListDataAutomationLibraryEntities` | data-automation-library | 0 |
| `bedrock:ListDataAutomationLibraryIngestionJobs` | data-automation-library | 0 |
| `bedrock:ListDataAutomationProjects` | blueprint, data-automation-project | 0 |
| `bedrock:ListDataSources` | knowledge-base | 0 |
| `bedrock:ListEnforcedGuardrailsConfiguration` | * | 0 |
| `bedrock:ListEvaluationJobs` | * | 0 |
| `bedrock:ListFlowAliases` | flow | 0 |
| `bedrock:ListFlowExecutionEvents` | flow, flow-alias, flow-execution | 0 |
| `bedrock:ListFlowExecutions` | flow, flow-alias | 0 |
| `bedrock:ListFlowVersions` | flow | 0 |
| `bedrock:ListFlows` | * | 0 |
| `bedrock:ListFoundationModelAgreementOffers` | * | 0 |
| `bedrock:ListFoundationModels` | * | 0 |
| `bedrock:ListGuardrails` | guardrail | 0 |
| `bedrock:ListImportedModels` | * | 0 |
| `bedrock:ListInferenceProfiles` | * | 0 |
| `bedrock:ListIngestionJobs` | knowledge-base | 0 |
| `bedrock:ListInvocationSteps` | session | 0 |
| `bedrock:ListInvocations` | session | 0 |
| `bedrock:ListKnowledgeBaseDocuments` | knowledge-base | 0 |
| `bedrock:ListKnowledgeBases` | * | 0 |
| `bedrock:ListModelCopyJobs` | * | 0 |
| `bedrock:ListModelCustomizationJobs` | * | 0 |
| `bedrock:ListModelEvaluationJobs` | * | 0 |
| `bedrock:ListModelImportJobs` | * | 0 |
| `bedrock:ListModelInvocationJobs` | * | 0 |
| `bedrock:ListPromptRouters` | * | 0 |
| `bedrock:ListPrompts` | prompt | 0 |
| `bedrock:ListProvisionedModelThroughputs` | * | 0 |
| `bedrock:ListRetrievalInstructionVersions` | retrieval-instruction | 0 |
| `bedrock:ListRetrievalInstructions` | * | 0 |
| `bedrock:ListSessions` | * | 0 |

List actions of Bedrock, fetched 16 Nov 2026.

## Service condition keys: 13

`aws:RequestTag/${TagKey}`, `aws:ResourceTag/${TagKey}`, `aws:TagKeys`, `bedrock:BearerTokenType`, `bedrock:DataRetentionMode`, `bedrock:GuardrailIdentifier`, `bedrock:InferenceProfileArn`, `bedrock:InlineAgentName`, `bedrock:ModelArn`, `bedrock:ProjectArn`, `bedrock:PromptRouterArn`, `bedrock:ServiceTier`, `bedrock:ThirdPartyKnowledgeBaseCredentialsSecretArn`

## Related pages

- [SageMaker IAM actions and access levels](/actions/sagemaker/)
- [CodeBuild IAM actions and access levels](/actions/codebuild/)
- [CodePipeline IAM actions and access levels](/actions/codepipeline/)

## Sources

- AWS Service Authorization Reference, programmatic access (fetched 16 Nov 2026): https://docs.aws.amazon.com/service-authorization/latest/reference/service-reference.html (read 16 Nov 2026)
- Bedrock service reference (JSON): https://servicereference.us-east-1.amazonaws.com/v1/bedrock/bedrock.json (read 16 Nov 2026)

Last reviewed 16 Nov 2026. Canonical: https://iampolicygenerator.com/actions/bedrock/

This site is not affiliated with Amazon Web Services. AWS service and product names are used only to describe AWS services.
