awsbedrockalpha

package module
v2.252.0-alpha.0 Latest Latest
Warning

This package is not in the latest version of its module.

 Go to latest Published: Apr 30, 2026 License: Apache-2.0 Imports: 17 Imported by: 1

README

Amazon Bedrock Construct Library

---

The APIs of higher level constructs in this module are experimental and under active development. They are subject to non-backward compatible changes or removal in any future version. These are not subject to the Semantic Versioning model and breaking changes will be announced in the release notes. This means that while you may use them, you may need to update your source code when upgrading to a newer version of this package.

Language Package
Typescript Logo TypeScript @aws-cdk/aws-bedrock-alpha

Amazon Bedrock is a fully managed service that offers a choice of high-performing foundation models (FMs) from leading AI companies and Amazon through a single API, along with a broad set of capabilities you need to build generative AI applications with security, privacy, and responsible AI.

This construct library facilitates the deployment of Bedrock Agents, enabling you to create sophisticated AI applications that can interact with your systems and data sources.

Table of contents

Agents

Amazon Bedrock Agents allow generative AI applications to automate complex, multistep tasks by seamlessly integrating with your company's systems, APIs, and data sources. It uses the reasoning of foundation models (FMs), APIs, and data to break down user requests, gather relevant information, and efficiently complete tasks.

Create an Agent

Building an agent is straightforward and fast. The following example creates an Agent with a simple instruction and default prompts:

agent := bedrock.NewAgent(this, jsii.String("Agent"), &AgentProps{
	FoundationModel: bedrock.BedrockFoundationModel_ANTHROPIC_CLAUDE_HAIKU_V1_0(),
	Instruction: jsii.String("You are a helpful and friendly agent that answers questions about literature."),
})

You can also create an agent with a guardrail:

// Create a guardrail to filter inappropriate content
guardrail := bedrock.NewGuardrail(this, jsii.String("bedrockGuardrails"), &GuardrailProps{
	GuardrailName: jsii.String("my-BedrockGuardrails"),
	Description: jsii.String("Legal ethical guardrails."),
})

guardrail.AddContentFilter(&ContentFilter{
	Type: bedrock.ContentFilterType_SEXUAL,
	InputStrength: bedrock.ContentFilterStrength_HIGH,
	OutputStrength: bedrock.ContentFilterStrength_MEDIUM,
})

// Create an agent with the guardrail
agentWithGuardrail := bedrock.NewAgent(this, jsii.String("AgentWithGuardrail"), &AgentProps{
	FoundationModel: bedrock.BedrockFoundationModel_ANTHROPIC_CLAUDE_HAIKU_V1_0(),
	Instruction: jsii.String("You are a helpful and friendly agent that answers questions about literature."),
	Guardrail: guardrail,
})
Agent Properties

The Bedrock Agent class supports the following properties.

Name Type Required Description
name string No The name of the agent. Defaults to a name generated by CDK
instruction string Yes The instruction used by the agent that determines how it will perform its task. Must have a minimum of 40 characters
foundationModel IBedrockInvokable Yes The foundation model used for orchestration by the agent
existingRole iam.IRole No The existing IAM Role for the agent to use. Must have a trust policy allowing Bedrock service to assume the role. Defaults to a new created role
shouldPrepareAgent boolean No Specifies whether to automatically update the DRAFT version of the agent after making changes. Defaults to false
idleSessionTTL Duration No How long sessions should be kept open for the agent. Session expires if no conversation occurs during this time. Defaults to 1 hour
kmsKey kms.IKey No The KMS key of the agent if custom encryption is configured. Defaults to AWS managed key
description string No A description of the agent. Defaults to no description
actionGroups AgentActionGroup[] No The Action Groups associated with the agent
guardrail IGuardrail No The guardrail that will be associated with the agent. Defaults to no guardrail
memory Memory No The type and configuration of the memory to maintain context across multiple sessions and recall past interactions. Defaults to no memory
promptOverrideConfiguration PromptOverrideConfiguration No Overrides some prompt templates in different parts of an agent sequence configuration
userInputEnabled boolean No Select whether the agent can prompt additional information from the user when it lacks enough information. Defaults to false
codeInterpreterEnabled boolean No Select whether the agent can generate, run, and troubleshoot code when trying to complete a task. Defaults to false
forceDelete boolean No Whether to delete the resource even if it's in use. Defaults to true
agentCollaboration AgentCollaboration No Configuration for agent collaboration settings, including type and collaborators. This property allows you to define how the agent collaborates with other agents and what collaborators it can work with. Defaults to no agent collaboration configuration
customOrchestrationExecutor CustomOrchestrationExecutor No The Lambda function to use for custom orchestration. If provided, orchestrationType is set to CUSTOM_ORCHESTRATION. If not provided, orchestrationType defaults to DEFAULT. Defaults to default orchestration
Action Groups

An action group defines functions your agent can call. The functions are Lambda functions. The action group uses an OpenAPI schema to tell the agent what your functions do and how to call them.

Action Group Properties

The AgentActionGroup class supports the following properties.

Name Type Required Description
name string No The name of the action group. Defaults to a name generated in the format 'action_group_quick_start_UUID'
description string No A description of the action group
apiSchema ApiSchema No The OpenAPI schema that defines the functions in the action group
executor ActionGroupExecutor No The Lambda function that executes the actions in the group
enabled boolean No Whether the action group is enabled. Defaults to true
forceDelete boolean No Whether to delete the resource even if it's in use. Defaults to false
functionSchema FunctionSchema No Defines functions that each define parameters that the agent needs to invoke from the user
parentActionGroupSignature ParentActionGroupSignature No The AWS Defined signature for enabling certain capabilities in your agent

There are three ways to provide an API schema for your action group:

From a local asset file (requires binding to scope):

actionGroupFunction := lambda.NewFunction(this, jsii.String("ActionGroupFunction"), &FunctionProps{
	Runtime: lambda.Runtime_PYTHON_3_12(),
	Handler: jsii.String("index.handler"),
	Code: lambda.Code_FromAsset(path.join(__dirname, jsii.String("../lambda/action-group"))),
})

// When using ApiSchema.fromLocalAsset, you must bind the schema to a scope
schema := bedrock.ApiSchema_FromLocalAsset(path.join(__dirname, jsii.String("action-group.yaml")))
schema.Bind(this)

actionGroup := bedrock.NewAgentActionGroup(&AgentActionGroupProps{
	Name: jsii.String("query-library"),
	Description: jsii.String("Use these functions to get information about the books in the library."),
	Executor: bedrock.ActionGroupExecutor_FromLambda(actionGroupFunction),
	Enabled: jsii.Boolean(true),
	ApiSchema: schema,
})

agent := bedrock.NewAgent(this, jsii.String("Agent"), &AgentProps{
	FoundationModel: bedrock.BedrockFoundationModel_ANTHROPIC_CLAUDE_HAIKU_V1_0(),
	Instruction: jsii.String("You are a helpful and friendly agent that answers questions about literature."),
})

agent.AddActionGroup(actionGroup)

From an inline OpenAPI schema:

inlineSchema := bedrock.ApiSchema_FromInline(jsii.String(`
openapi: 3.0.3
info:
  title: Library API
  version: 1.0.0
paths:
  /search:
    get:
      summary: Search for books
      operationId: searchBooks
      parameters:
        - name: query
          in: query
          required: true
          schema:
            type: string
`))

actionGroupFunction := lambda.NewFunction(this, jsii.String("ActionGroupFunction"), &FunctionProps{
	Runtime: lambda.Runtime_PYTHON_3_12(),
	Handler: jsii.String("index.handler"),
	Code: lambda.Code_FromAsset(path.join(__dirname, jsii.String("../lambda/action-group"))),
})

actionGroup := bedrock.NewAgentActionGroup(&AgentActionGroupProps{
	Name: jsii.String("query-library"),
	Description: jsii.String("Use these functions to get information about the books in the library."),
	Executor: bedrock.ActionGroupExecutor_FromLambda(actionGroupFunction),
	Enabled: jsii.Boolean(true),
	ApiSchema: inlineSchema,
})

agent := bedrock.NewAgent(this, jsii.String("Agent"), &AgentProps{
	FoundationModel: bedrock.BedrockFoundationModel_ANTHROPIC_CLAUDE_HAIKU_V1_0(),
	Instruction: jsii.String("You are a helpful and friendly agent that answers questions about literature."),
})

agent.AddActionGroup(actionGroup)

From an existing S3 file:

bucket := s3.Bucket_FromBucketName(this, jsii.String("ExistingBucket"), jsii.String("my-schema-bucket"))
s3Schema := bedrock.ApiSchema_FromS3File(bucket, jsii.String("schemas/action-group.yaml"))

actionGroupFunction := lambda.NewFunction(this, jsii.String("ActionGroupFunction"), &FunctionProps{
	Runtime: lambda.Runtime_PYTHON_3_12(),
	Handler: jsii.String("index.handler"),
	Code: lambda.Code_FromAsset(path.join(__dirname, jsii.String("../lambda/action-group"))),
})

actionGroup := bedrock.NewAgentActionGroup(&AgentActionGroupProps{
	Name: jsii.String("query-library"),
	Description: jsii.String("Use these functions to get information about the books in the library."),
	Executor: bedrock.ActionGroupExecutor_FromLambda(actionGroupFunction),
	Enabled: jsii.Boolean(true),
	ApiSchema: s3Schema,
})

agent := bedrock.NewAgent(this, jsii.String("Agent"), &AgentProps{
	FoundationModel: bedrock.BedrockFoundationModel_ANTHROPIC_CLAUDE_HAIKU_V1_0(),
	Instruction: jsii.String("You are a helpful and friendly agent that answers questions about literature."),
})

agent.AddActionGroup(actionGroup)
Using FunctionSchema with Action Groups

As an alternative to using OpenAPI schemas, you can define functions directly using the FunctionSchema class. This approach provides a more structured way to define the functions that your agent can call.

actionGroupFunction := lambda.NewFunction(this, jsii.String("ActionGroupFunction"), &FunctionProps{
	Runtime: lambda.Runtime_PYTHON_3_12(),
	Handler: jsii.String("index.handler"),
	Code: lambda.Code_FromAsset(path.join(__dirname, jsii.String("../lambda/action-group"))),
})

// Define a function schema with parameters
functionSchema := bedrock.NewFunctionSchema(&FunctionSchemaProps{
	Functions: []FunctionProps{
		&FunctionProps{
			Name: jsii.String("searchBooks"),
			Description: jsii.String("Search for books in the library catalog"),
			Parameters: map[string]FunctionParameterProps{
				"query": &FunctionParameterProps{
					"type": bedrock.ParameterType_STRING,
					"required": jsii.Boolean(true),
					"description": jsii.String("The search query string"),
				},
				"maxResults": &FunctionParameterProps{
					"type": bedrock.ParameterType_INTEGER,
					"required": jsii.Boolean(false),
					"description": jsii.String("Maximum number of results to return"),
				},
				"includeOutOfPrint": &FunctionParameterProps{
					"type": bedrock.ParameterType_BOOLEAN,
					"required": jsii.Boolean(false),
					"description": jsii.String("Whether to include out-of-print books"),
				},
			},
			RequireConfirmation: bedrock.RequireConfirmation_DISABLED,
		},
		&FunctionProps{
			Name: jsii.String("getBookDetails"),
			Description: jsii.String("Get detailed information about a specific book"),
			Parameters: map[string]FunctionParameterProps{
				"bookId": &FunctionParameterProps{
					"type": bedrock.ParameterType_STRING,
					"required": jsii.Boolean(true),
					"description": jsii.String("The unique identifier of the book"),
				},
			},
			RequireConfirmation: bedrock.RequireConfirmation_ENABLED,
		},
	},
})

// Create an action group using the function schema
actionGroup := bedrock.NewAgentActionGroup(&AgentActionGroupProps{
	Name: jsii.String("library-functions"),
	Description: jsii.String("Functions for interacting with the library catalog"),
	Executor: bedrock.ActionGroupExecutor_FromLambda(actionGroupFunction),
	FunctionSchema: functionSchema,
	Enabled: jsii.Boolean(true),
})

agent := bedrock.NewAgent(this, jsii.String("Agent"), &AgentProps{
	FoundationModel: bedrock.BedrockFoundationModel_ANTHROPIC_CLAUDE_HAIKU_V1_0(),
	Instruction: jsii.String("You are a helpful and friendly agent that answers questions about literature."),
	ActionGroups: []AgentActionGroup{
		actionGroup,
	},
})

The FunctionSchema approach offers several advantages:

  • Type-safe definition of functions and parameters
  • Built-in validation of parameter names, descriptions, and other properties
  • Clear structure that maps directly to the AWS Bedrock API
  • Support for parameter types including string, number, integer, boolean, array, and object
  • Option to require user confirmation before executing specific functions

If you chose to load your schema file from S3, the construct will provide the necessary permissions to your agent's execution role to access the schema file from the specific bucket. Similar to performing the operation through the console, the agent execution role will get a permission like:

{
    "Version": "2012-10-17",
    "Statement": [
        {
            "Sid": "AmazonBedrockAgentS3PolicyProd",
            "Effect": "Allow",
            "Action": [
                "s3:GetObject"
            ],
            "Resource": [
                "arn:aws:s3:::<BUCKET_NAME>/<OBJECT_KEY>"
            ],
            "Condition": {
                "StringEquals": {
                    "aws:ResourceAccount": "ACCOUNT_NUMBER"
                }
            }
        }
    ]
}

// create a bucket containing the input schema
schemaBucket := s3.NewBucket(this, jsii.String("SchemaBucket"), &BucketProps{
	EnforceSSL: jsii.Boolean(true),
	Versioned: jsii.Boolean(true),
	PublicReadAccess: jsii.Boolean(false),
	BlockPublicAccess: s3.BlockPublicAccess_BLOCK_ALL(),
	Encryption: s3.BucketEncryption_S3_MANAGED,
	RemovalPolicy: awscdk.RemovalPolicy_DESTROY,
	AutoDeleteObjects: jsii.Boolean(true),
})

// deploy the local schema file to S3
deployement := awscdk.Aws_s3_deployment.NewBucketDeployment(this, jsii.String("DeployWebsite"), &BucketDeploymentProps{
	Sources: []ISource{
		awscdk.*Aws_s3_deployment.Source_Asset(path.join(__dirname, jsii.String("../inputschema"))),
	},
	DestinationBucket: schemaBucket,
	DestinationKeyPrefix: jsii.String("inputschema"),
})

// create the agent
agent := bedrock.NewAgent(this, jsii.String("Agent"), &AgentProps{
	FoundationModel: bedrock.BedrockFoundationModel_ANTHROPIC_CLAUDE_3_5_SONNET_V1_0(),
	Instruction: jsii.String("You are a helpful and friendly agent that answers questions about literature."),
	UserInputEnabled: jsii.Boolean(true),
	ShouldPrepareAgent: jsii.Boolean(true),
})

// create a lambda function
actionGroupFunction := lambda.NewFunction(this, jsii.String("ActionGroupFunction"), &FunctionProps{
	Runtime: lambda.Runtime_PYTHON_3_12(),
	Handler: jsii.String("index.handler"),
	Code: lambda.Code_FromAsset(path.join(__dirname, jsii.String("../lambda/action-group"))),
})

// create an action group and read the schema file from S3
actionGroup := bedrock.NewAgentActionGroup(&AgentActionGroupProps{
	Name: jsii.String("query-library"),
	Description: jsii.String("Use these functions to get information about the books in the library."),
	Executor: bedrock.ActionGroupExecutor_FromLambda(actionGroupFunction),
	Enabled: jsii.Boolean(true),
	ApiSchema: bedrock.ApiSchema_FromS3File(schemaBucket, jsii.String("inputschema/action-group.yaml")),
})

// add the action group to the agent
agent.AddActionGroup(actionGroup)

// add dependency for the agent on the s3 deployment
agent.Node.AddDependency(deployement)
Prepare the Agent

The Agent constructs take an optional parameter shouldPrepareAgent to indicate that the Agent should be prepared after any updates to an agent or action group. This may increase the time to create and update those resources. By default, this value is false.

Prepare Agent Properties
Name Type Required Description
shouldPrepareAgent boolean No Whether to automatically update the DRAFT version of the agent after making changes. Defaults to false

Creating an agent alias will not prepare the agent, so if you create an alias using the AgentAlias resource then you should set shouldPrepareAgent to true.

Prompt Override Configuration

Bedrock Agents allows you to customize the prompts and LLM configuration for different steps in the agent sequence. The implementation provides type-safe configurations for each step type, ensuring correct usage at compile time.

Prompt Override Configuration Properties
Name Type Required Description
steps PromptStepConfiguration[] Yes Array of step configurations for different parts of the agent sequence
parser lambda.IFunction No Lambda function for custom parsing of agent responses
Prompt Step Configuration Properties

Each step in the steps array supports the following properties:

Name Type Required Description
stepType AgentStepType Yes The type of step being configured (PRE_PROCESSING, ORCHESTRATION, POST_PROCESSING, ROUTING_CLASSIFIER, MEMORY_SUMMARIZATION, KNOWLEDGE_BASE_RESPONSE_GENERATION)
stepEnabled boolean No Whether this step is enabled. Defaults to true
customPromptTemplate string No Custom prompt template to use for this step
inferenceConfig InferenceConfiguration No Configuration for model inference parameters
foundationModel BedrockFoundationModel No Alternative foundation model to use for this step (only valid for ROUTING_CLASSIFIER step)
useCustomParser boolean No Whether to use a custom parser for this step. Requires parser to be provided in PromptOverrideConfiguration
Inference Configuration Properties

When providing inferenceConfig, the following properties are supported:

Name Type Required Description
temperature number No Controls randomness in the model's output (0.0-1.0)
topP number No Controls diversity via nucleus sampling (0.0-1.0)
topK number No Controls diversity by limiting the cumulative probability
maximumLength number No Maximum length of generated text
stopSequences string[] No Sequences where the model should stop generating

The following steps can be configured:

  • PRE_PROCESSING: Prepares the user input for orchestration
  • ORCHESTRATION: Main step that determines the agent's actions
  • POST_PROCESSING: Refines the agent's response
  • ROUTING_CLASSIFIER: Classifies and routes requests to appropriate collaborators
  • MEMORY_SUMMARIZATION: Summarizes conversation history for memory retention
  • KNOWLEDGE_BASE_RESPONSE_GENERATION: Generates responses using knowledge base content

Example with pre-processing configuration:

agent := bedrock.NewAgent(this, jsii.String("Agent"), &AgentProps{
	FoundationModel: bedrock.BedrockFoundationModel_AMAZON_NOVA_LITE_V1(),
	Instruction: jsii.String("You are a helpful assistant."),
	PromptOverrideConfiguration: bedrock.PromptOverrideConfiguration_FromSteps([]PromptStepConfigBase{
		&PromptStepConfigBase{
			StepType: bedrock.AgentStepType_PRE_PROCESSING,
			StepEnabled: jsii.Boolean(true),
			CustomPromptTemplate: jsii.String("Your custom prompt template here"),
			InferenceConfig: &InferenceConfiguration{
				Temperature: jsii.Number(0),
				TopP: jsii.Number(1),
				TopK: jsii.Number(250),
				MaximumLength: jsii.Number(1),
				StopSequences: []*string{
					jsii.String("\n\nHuman:"),
				},
			},
		},
	}),
})

Example with routing classifier and foundation model:

agent := bedrock.NewAgent(this, jsii.String("Agent"), &AgentProps{
	FoundationModel: bedrock.BedrockFoundationModel_AMAZON_NOVA_LITE_V1(),
	Instruction: jsii.String("You are a helpful assistant."),
	PromptOverrideConfiguration: bedrock.PromptOverrideConfiguration_FromSteps([]PromptStepConfigBase{
		&PromptRoutingClassifierConfigCustomParser{
			StepType: bedrock.AgentStepType_ROUTING_CLASSIFIER,
			StepEnabled: jsii.Boolean(true),
			CustomPromptTemplate: jsii.String("Your routing template here"),
			FoundationModel: bedrock.BedrockFoundationModel_ANTHROPIC_CLAUDE_V2(),
		}.(PromptRoutingClassifierConfigCustomParser),
	}),
})

Using a custom Lambda parser:

parserFunction := lambda.NewFunction(this, jsii.String("ParserFunction"), &FunctionProps{
	Runtime: lambda.Runtime_PYTHON_3_10(),
	Handler: jsii.String("index.handler"),
	Code: lambda.Code_FromAsset(jsii.String("lambda")),
})

agent := bedrock.NewAgent(this, jsii.String("Agent"), &AgentProps{
	FoundationModel: bedrock.BedrockFoundationModel_AMAZON_NOVA_LITE_V1(),
	Instruction: jsii.String("You are a helpful assistant."),
	PromptOverrideConfiguration: bedrock.PromptOverrideConfiguration_WithCustomParser(&CustomParserProps{
		Parser: parserFunction,
		PreProcessingStep: &PromptPreProcessingConfigCustomParser{
			StepType: bedrock.AgentStepType_PRE_PROCESSING,
			UseCustomParser: jsii.Boolean(true),
		},
	}),
})

Foundation models can only be specified for the ROUTING_CLASSIFIER step.

Memory Configuration

Agents can maintain context across multiple sessions and recall past interactions using memory. This feature is useful for creating a more coherent conversational experience.

Memory Configuration Properties
Name Type Required Description
maxRecentSessions number No Maximum number of recent session summaries to retain
memoryDuration Duration No How long to retain session summaries

Example:

agent := bedrock.NewAgent(this, jsii.String("MyAgent"), &AgentProps{
	AgentName: jsii.String("MyAgent"),
	Instruction: jsii.String("Your instruction here"),
	FoundationModel: bedrock.BedrockFoundationModel_AMAZON_NOVA_LITE_V1(),
	Memory: awsbedrockalpha.Memory_SessionSummary(&SessionSummaryMemoryProps{
		MaxRecentSessions: jsii.Number(10),
		 // Keep the last 10 session summaries
		MemoryDuration: awscdk.Duration_Days(jsii.Number(20)),
	}),
})
Agent Collaboration

Agent Collaboration enables multiple Bedrock Agents to work together on complex tasks. This feature allows agents to specialize in different areas and collaborate to provide more comprehensive responses to user queries.

Agent Collaboration Properties
Name Type Required Description
type AgentCollaboratorType Yes Type of collaboration (SUPERVISOR or PEER)
collaborators AgentCollaborator[] Yes List of agent collaborators
Agent Collaborator Properties
Name Type Required Description
agentAlias AgentAlias Yes The agent alias to collaborate with
collaborationInstruction string Yes Instructions for how to collaborate with this agent
collaboratorName string Yes Name of the collaborator
relayConversationHistory boolean No Whether to relay conversation history to the collaborator. Defaults to false

Example:

// Create a specialized agent
customerSupportAgent := bedrock.NewAgent(this, jsii.String("CustomerSupportAgent"), &AgentProps{
	Instruction: jsii.String("You specialize in answering customer support questions."),
	FoundationModel: bedrock.BedrockFoundationModel_AMAZON_NOVA_LITE_V1(),
})

// Create an agent alias
customerSupportAlias := bedrock.NewAgentAlias(this, jsii.String("CustomerSupportAlias"), &AgentAliasProps{
	Agent: customerSupportAgent,
	AgentAliasName: jsii.String("production"),
})

// Create a main agent that collaborates with the specialized agent
mainAgent := bedrock.NewAgent(this, jsii.String("MainAgent"), &AgentProps{
	Instruction: jsii.String("You route specialized questions to other agents."),
	FoundationModel: bedrock.BedrockFoundationModel_AMAZON_NOVA_LITE_V1(),
	AgentCollaboration: map[string]interface{}{
		"type": bedrock.AgentCollaboratorType_SUPERVISOR,
		"collaborators": []AgentCollaborator{
			bedrock.NewAgentCollaborator(&AgentCollaboratorProps{
				"agentAlias": customerSupportAlias,
				"collaborationInstruction": jsii.String("Route customer support questions to this agent."),
				"collaboratorName": jsii.String("CustomerSupport"),
				"relayConversationHistory": jsii.Boolean(true),
			}),
		},
	},
})
Custom Orchestration

Custom Orchestration allows you to override the default agent orchestration flow with your own Lambda function. This enables more control over how the agent processes user inputs and invokes action groups.

When you provide a customOrchestrationExecutor, the agent's orchestrationType is automatically set to CUSTOM_ORCHESTRATION. If no customOrchestrationExecutor is provided, the orchestrationType defaults to DEFAULT, using Amazon Bedrock's built-in orchestration.

Custom Orchestration Properties
Name Type Required Description
function lambda.IFunction Yes The Lambda function that implements the custom orchestration logic

Example:

orchestrationFunction := lambda.NewFunction(this, jsii.String("OrchestrationFunction"), &FunctionProps{
	Runtime: lambda.Runtime_PYTHON_3_10(),
	Handler: jsii.String("index.handler"),
	Code: lambda.Code_FromAsset(jsii.String("lambda/orchestration")),
})

agent := bedrock.NewAgent(this, jsii.String("CustomOrchestrationAgent"), &AgentProps{
	Instruction: jsii.String("You are a helpful assistant with custom orchestration logic."),
	FoundationModel: bedrock.BedrockFoundationModel_AMAZON_NOVA_LITE_V1(),
	CustomOrchestrationExecutor: bedrock.CustomOrchestrationExecutor_FromLambda(orchestrationFunction),
})
Agent Alias

After you have sufficiently iterated on your working draft and are satisfied with the behavior of your agent, you can set it up for deployment and integration into your application by creating aliases.

To deploy your agent, you need to create an alias. During alias creation, Amazon Bedrock automatically creates a version of your agent. The alias points to this newly created version. You can point the alias to a previously created version if necessary. You then configure your application to make API calls to that alias.

By default, the Agent resource creates a test alias named 'AgentTestAlias' that points to the 'DRAFT' version. This test alias is accessible via the testAlias property of the agent. You can also create additional aliases for different environments using the AgentAlias construct.

Agent Alias Properties
Name Type Required Description
agent Agent Yes The agent to create an alias for
agentAliasName string No The name of the agent alias. Defaults to a name generated by CDK
description string No A description of the agent alias. Defaults to no description
routingConfiguration AgentAliasRoutingConfiguration No Configuration for routing traffic between agent versions
agentVersion string No The version of the agent to use. If not specified, a new version is created

When redeploying an agent with changes, you must ensure the agent version is updated to avoid deployment failures with "agent already exists" errors. The recommended way to handle this is to include the lastUpdated property in the agent's description, which automatically updates whenever the agent is modified. This ensures a new version is created on each deployment.

Example:

agent := bedrock.NewAgent(this, jsii.String("Agent"), &AgentProps{
	FoundationModel: bedrock.BedrockFoundationModel_ANTHROPIC_CLAUDE_HAIKU_V1_0(),
	Instruction: jsii.String("You are a helpful and friendly agent that answers questions about literature."),
})

agentAlias := bedrock.NewAgentAlias(this, jsii.String("myAlias"), &AgentAliasProps{
	AgentAliasName: jsii.String("production"),
	Agent: agent,
	Description: fmt.Sprintf("Production version of my agent. Created at %v", agent.LastUpdated),
})

Guardrails

Amazon Bedrock's Guardrails feature enables you to implement robust governance and control mechanisms for your generative AI applications, ensuring alignment with your specific use cases and responsible AI policies. Guardrails empowers you to create multiple tailored policy configurations, each designed to address the unique requirements and constraints of different use cases. These policy configurations can then be seamlessly applied across multiple foundation models (FMs) and Agents, ensuring a consistent user experience and standardizing safety, security, and privacy controls throughout your generative AI ecosystem.

With Guardrails, you can define and enforce granular, customizable policies to precisely govern the behavior of your generative AI applications. You can configure the following policies in a guardrail to avoid undesirable and harmful content and remove sensitive information for privacy protection.

Content filters – Adjust filter strengths to block input prompts or model responses containing harmful content. Denied topics – Define a set of topics that are undesirable in the context of your application. These topics will be blocked if detected in user queries or model responses. Word filters – Configure filters to block undesirable words, phrases, and profanity. Such words can include offensive terms, competitor names etc. Sensitive information filters – Block or mask sensitive information such as personally identifiable information (PII) or custom regex in user inputs and model responses. You can create a Guardrail with a minimum blockedInputMessaging, blockedOutputsMessaging and default content filter policy.

Basic Guardrail Creation
TypeScript
guardrail := bedrock.NewGuardrail(this, jsii.String("bedrockGuardrails"), &GuardrailProps{
	GuardrailName: jsii.String("my-BedrockGuardrails"),
	Description: jsii.String("Legal ethical guardrails."),
})
Guardrail Properties
Property Type Required Description
guardrailName string Yes The name of the guardrail
description string No The description of the guardrail
blockedInputMessaging string No The message to return when the guardrail blocks a prompt. Default: "Sorry, your query violates our usage policy."
blockedOutputsMessaging string No The message to return when the guardrail blocks a model response. Default: "Sorry, I am unable to answer your question because of our usage policy."
kmsKey IKey No A custom KMS key to use for encrypting data. Default: Your data is encrypted by default with a key that AWS owns and manages for you.
crossRegionConfig GuardrailCrossRegionConfigProperty No The cross-region configuration for the guardrail. This enables cross-region inference for enhanced language support and filtering capabilities. Default: No cross-region configuration
contentFilters ContentFilter[] No The content filters to apply to the guardrail
contentFiltersTierConfig TierConfig No The tier configuration to apply to content filters. Default: TierConfig.CLASSIC
deniedTopics Topic[] No Up to 30 denied topics to block user inputs or model responses associated with the topic
topicsTierConfig TierConfig No The tier configuration to apply to topic filters. Default: TierConfig.CLASSIC
wordFilters string[] No The word filters to apply to the guardrail
managedWordListFilters ManagedWordFilterType[] No The managed word filters to apply to the guardrail
piiFilters PIIFilter[] No The PII filters to apply to the guardrail
regexFilters RegexFilter[] No The regular expression (regex) filters to apply to the guardrail
contextualGroundingFilters ContextualGroundingFilter[] No The contextual grounding filters to apply to the guardrail
Filter Types
Content Filters

Content filters allow you to block input prompts or model responses containing harmful content. You can adjust the filter strength and configure separate actions for input and output.

Content Filter Configuration
guardrail := bedrock.NewGuardrail(this, jsii.String("bedrockGuardrails"), &GuardrailProps{
	GuardrailName: jsii.String("my-BedrockGuardrails"),
	// Configure tier for content filters (optional)
	ContentFiltersTierConfig: bedrock.TierConfig_STANDARD,
})

guardrail.AddContentFilter(&ContentFilter{
	Type: bedrock.ContentFilterType_SEXUAL,
	InputStrength: bedrock.ContentFilterStrength_HIGH,
	OutputStrength: bedrock.ContentFilterStrength_MEDIUM,
	// props below are optional
	InputAction: bedrock.GuardrailAction_BLOCK,
	InputEnabled: jsii.Boolean(true),
	OutputAction: bedrock.GuardrailAction_NONE,
	OutputEnabled: jsii.Boolean(true),
	InputModalities: []ModalityType{
		bedrock.ModalityType_TEXT,
		bedrock.ModalityType_IMAGE,
	},
	OutputModalities: []ModalityType{
		bedrock.ModalityType_TEXT,
	},
})

Available content filter types:

  • SEXUAL: Describes input prompts and model responses that indicates sexual interest, activity, or arousal
  • VIOLENCE: Describes input prompts and model responses that includes glorification of or threats to inflict physical pain
  • HATE: Describes input prompts and model responses that discriminate, criticize, insult, denounce, or dehumanize a person or group
  • INSULTS: Describes input prompts and model responses that includes demeaning, humiliating, mocking, insulting, or belittling language
  • MISCONDUCT: Describes input prompts and model responses that seeks or provides information about engaging in misconduct activity
  • PROMPT_ATTACK: Enable to detect and block user inputs attempting to override system instructions

Available content filter strengths:

  • NONE: No filtering
  • LOW: Light filtering
  • MEDIUM: Moderate filtering
  • HIGH: Strict filtering

Available guardrail actions:

  • BLOCK: Blocks the content from being processed
  • ANONYMIZE: Masks the content with an identifier tag
  • NONE: Takes no action

Warning: the ANONYMIZE action is not available in all configurations. Please refer to the documentation of each filter to see which ones support

Available modality types:

  • TEXT: Text modality for content filters
  • IMAGE: Image modality for content filters
Tier Configuration

Guardrails support tier configurations that determine the level of language support and robustness for content and topic filters. You can configure separate tier settings for content filters and topic filters.

Tier Configuration Options
guardrail := bedrock.NewGuardrail(this, jsii.String("bedrockGuardrails"), &GuardrailProps{
	GuardrailName: jsii.String("my-BedrockGuardrails"),
	// Configure tier for content filters
	ContentFiltersTierConfig: bedrock.TierConfig_STANDARD,
	// Configure tier for topic filters
	TopicsTierConfig: bedrock.TierConfig_CLASSIC,
})

Available tier configurations:

  • CLASSIC: Provides established guardrails functionality supporting English, French, and Spanish languages
  • STANDARD: Provides a more robust solution than the CLASSIC tier and has more comprehensive language support. This tier requires that your guardrail use cross-Region inference

Note: The STANDARD tier provides enhanced language support and more comprehensive filtering capabilities, but requires cross-Region inference to be enabled for your guardrail.

Cross-Region Configuration

You can configure a system-defined guardrail profile to use with your guardrail. Guardrail profiles define the destination AWS Regions where guardrail inference requests can be automatically routed. Using guardrail profiles helps maintain guardrail performance and reliability when demand increases.

Cross-Region Configuration Properties
Property Type Required Description
guardrailProfileArn string Yes The ARN of the system-defined guardrail profile that defines the destination AWS Regions where guardrail inference requests can be automatically routed
Cross-Region Configuration Example
guardrail := bedrock.NewGuardrail(this, jsii.String("bedrockGuardrails"), &GuardrailProps{
	GuardrailName: jsii.String("my-BedrockGuardrails"),
	Description: jsii.String("Guardrail with cross-region configuration for enhanced language support"),
	CrossRegionConfig: &GuardrailCrossRegionConfigProperty{
		GuardrailProfileArn: jsii.String("arn:aws:bedrock:us-east-1:123456789012:guardrail-profile/my-profile"),
	},
	// Use STANDARD tier for enhanced capabilities
	ContentFiltersTierConfig: bedrock.TierConfig_STANDARD,
	TopicsTierConfig: bedrock.TierConfig_STANDARD,
})

Note: Cross-region configuration is required when using the STANDARD tier for content and topic filters. It helps maintain guardrail performance and reliability when demand increases by automatically routing inference requests to appropriate regions.

You will need to provide the necessary permissions for cross region: https://docs.aws.amazon.com/bedrock/latest/userguide/guardrail-profiles-permissions.html .

Denied Topics

Denied topics allow you to define a set of topics that are undesirable in the context of your application. These topics will be blocked if detected in user queries or model responses. You can configure separate actions for input and output.

Denied Topic Configuration
guardrail := bedrock.NewGuardrail(this, jsii.String("bedrockGuardrails"), &GuardrailProps{
	GuardrailName: jsii.String("my-BedrockGuardrails"),
	// Configure tier for topic filters (optional)
	TopicsTierConfig: bedrock.TierConfig_STANDARD,
})

// Use a predefined topic
guardrail.AddDeniedTopicFilter(bedrock.Topic_FINANCIAL_ADVICE())

// Create a custom topic with input/output actions
guardrail.AddDeniedTopicFilter(bedrock.Topic_Custom(&CustomTopicProps{
	Name: jsii.String("Legal_Advice"),
	Definition: jsii.String("Offering guidance or suggestions on legal matters, legal actions, interpretation of laws, or legal rights and responsibilities."),
	Examples: []*string{
		jsii.String("Can I sue someone for this?"),
		jsii.String("What are my legal rights in this situation?"),
		jsii.String("Is this action against the law?"),
		jsii.String("What should I do to file a legal complaint?"),
		jsii.String("Can you explain this law to me?"),
	},
	// props below are optional
	InputAction: bedrock.GuardrailAction_BLOCK,
	InputEnabled: jsii.Boolean(true),
	OutputAction: bedrock.GuardrailAction_NONE,
	OutputEnabled: jsii.Boolean(true),
}))
Word Filters

Word filters allow you to block specific words, phrases, or profanity in user inputs and model responses. You can configure separate actions for input and output.

Word Filter Configuration
guardrail := bedrock.NewGuardrail(this, jsii.String("bedrockGuardrails"), &GuardrailProps{
	GuardrailName: jsii.String("my-BedrockGuardrails"),
})

// Add managed word list with input/output actions
guardrail.AddManagedWordListFilter(&ManagedWordFilter{
	Type: bedrock.ManagedWordFilterType_PROFANITY,
	InputAction: bedrock.GuardrailAction_BLOCK,
	InputEnabled: jsii.Boolean(true),
	OutputAction: bedrock.GuardrailAction_NONE,
	OutputEnabled: jsii.Boolean(true),
})

// Add individual words
guardrail.AddWordFilter(&WordFilter{
	Text: jsii.String("drugs"),
})
guardrail.AddWordFilter(&WordFilter{
	Text: jsii.String("competitor"),
})

// Add words from a file
guardrail.AddWordFilterFromFile(jsii.String("./scripts/wordsPolicy.csv"))
PII Filters

PII filters allow you to detect and handle personally identifiable information in user inputs and model responses. You can configure separate actions for input and output.

The PII types are organized into enum-like classes for better type safety and transpilation compatibility:

  • GeneralPIIType: General PII types like addresses, emails, names, phone numbers
  • FinancePIIType: Financial information like credit card numbers, PINs, SWIFT codes
  • InformationTechnologyPIIType: IT-related data like URLs, IP addresses, AWS keys
  • USASpecificPIIType: US-specific identifiers like SSNs, passport numbers
  • CanadaSpecificPIIType: Canada-specific identifiers like health numbers, SINs
  • UKSpecificPIIType: UK-specific identifiers like NHS numbers, NI numbers
PII Filter Configuration
guardrail := bedrock.NewGuardrail(this, jsii.String("bedrockGuardrails"), &GuardrailProps{
	GuardrailName: jsii.String("my-BedrockGuardrails"),
})

// Add PII filter for addresses with input/output actions
guardrail.AddPIIFilter(&PIIFilter{
	Type: bedrock.GeneralPIIType_ADDRESS(),
	Action: bedrock.GuardrailAction_BLOCK,
	// below props are optional
	InputAction: bedrock.GuardrailAction_BLOCK,
	InputEnabled: jsii.Boolean(true),
	OutputAction: bedrock.GuardrailAction_ANONYMIZE,
	OutputEnabled: jsii.Boolean(true),
})

// Add PII filter for credit card numbers with input/output actions
guardrail.AddPIIFilter(&PIIFilter{
	Type: bedrock.FinancePIIType_CREDIT_DEBIT_CARD_NUMBER(),
	Action: bedrock.GuardrailAction_BLOCK,
	// below props are optional
	InputAction: bedrock.GuardrailAction_BLOCK,
	InputEnabled: jsii.Boolean(true),
	OutputAction: bedrock.GuardrailAction_ANONYMIZE,
	OutputEnabled: jsii.Boolean(true),
})

// Add PII filter for email addresses
guardrail.AddPIIFilter(&PIIFilter{
	Type: bedrock.GeneralPIIType_EMAIL(),
	Action: bedrock.GuardrailAction_ANONYMIZE,
})

// Add PII filter for US Social Security Numbers
guardrail.AddPIIFilter(&PIIFilter{
	Type: bedrock.USASpecificPIIType_US_SOCIAL_SECURITY_NUMBER(),
	Action: bedrock.GuardrailAction_BLOCK,
})

// Add PII filter for IP addresses
guardrail.AddPIIFilter(&PIIFilter{
	Type: bedrock.InformationTechnologyPIIType_IP_ADDRESS(),
	Action: bedrock.GuardrailAction_ANONYMIZE,
})
Available PII Types

GeneralPIIType:

  • ADDRESS: Physical addresses
  • AGE: Individual's age
  • DRIVER_ID: Driver's license numbers
  • EMAIL: Email addresses
  • LICENSE_PLATE: Vehicle license plates
  • NAME: Individual names
  • PASSWORD: Passwords
  • PHONE: Phone numbers
  • USERNAME: User account names
  • VEHICLE_IDENTIFICATION_NUMBER: Vehicle VINs

FinancePIIType:

  • CREDIT_DEBIT_CARD_CVV: Card verification codes
  • CREDIT_DEBIT_CARD_EXPIRY: Card expiration dates
  • CREDIT_DEBIT_CARD_NUMBER: Credit/debit card numbers
  • PIN: Personal identification numbers
  • SWIFT_CODE: Bank SWIFT codes
  • INTERNATIONAL_BANK_ACCOUNT_NUMBER: IBAN numbers

InformationTechnologyPIIType:

  • URL: Web addresses
  • IP_ADDRESS: IPv4 addresses
  • MAC_ADDRESS: Network interface MAC addresses
  • AWS_ACCESS_KEY: AWS access key IDs
  • AWS_SECRET_KEY: AWS secret access keys

USASpecificPIIType:

  • US_BANK_ACCOUNT_NUMBER: US bank account numbers
  • US_BANK_ROUTING_NUMBER: US bank routing numbers
  • US_INDIVIDUAL_TAX_IDENTIFICATION_NUMBER: US ITINs
  • US_PASSPORT_NUMBER: US passport numbers
  • US_SOCIAL_SECURITY_NUMBER: US Social Security Numbers

CanadaSpecificPIIType:

  • CA_HEALTH_NUMBER: Canadian Health Service Numbers
  • CA_SOCIAL_INSURANCE_NUMBER: Canadian Social Insurance Numbers

UKSpecificPIIType:

  • UK_NATIONAL_HEALTH_SERVICE_NUMBER: UK NHS numbers
  • UK_NATIONAL_INSURANCE_NUMBER: UK National Insurance numbers
  • UK_UNIQUE_TAXPAYER_REFERENCE_NUMBER: UK UTR numbers
Regex Filters

Regex filters allow you to detect and handle custom patterns in user inputs and model responses. You can configure separate actions for input and output.

Regex Filter Configuration
guardrail := bedrock.NewGuardrail(this, jsii.String("bedrockGuardrails"), &GuardrailProps{
	GuardrailName: jsii.String("my-BedrockGuardrails"),
})
// Add regex filter with input/output actions
guardrail.AddRegexFilter(&RegexFilter{
	Name: jsii.String("TestRegexFilter"),
	Pattern: jsii.String("test-pattern"),
	Action: bedrock.GuardrailAction_ANONYMIZE,
	// below props are optional
	Description: jsii.String("This is a test regex filter"),
	InputAction: bedrock.GuardrailAction_BLOCK,
	InputEnabled: jsii.Boolean(true),
	OutputAction: bedrock.GuardrailAction_ANONYMIZE,
	OutputEnabled: jsii.Boolean(true),
})
Contextual Grounding Filters

Contextual grounding filters allow you to ensure that model responses are factually correct and relevant to the user's query. You can configure the action and enable/disable the filter.

Contextual Grounding Filter Configuration
guardrail := bedrock.NewGuardrail(this, jsii.String("bedrockGuardrails"), &GuardrailProps{
	GuardrailName: jsii.String("my-BedrockGuardrails"),
})
// Add contextual grounding filter with action and enabled flag
guardrail.AddContextualGroundingFilter(&ContextualGroundingFilter{
	Type: bedrock.ContextualGroundingFilterType_GROUNDING,
	Threshold: jsii.Number(0.8),
	// the properties below are optional
	Action: bedrock.GuardrailAction_BLOCK,
	Enabled: jsii.Boolean(true),
})
Guardrail Methods
Method Description
addContentFilter() Adds a content filter to the guardrail
addDeniedTopicFilter() Adds a denied topic filter to the guardrail
addWordFilter() Adds a word filter to the guardrail
addManagedWordListFilter() Adds a managed word list filter to the guardrail
addWordFilterFromFile() Adds word filters from a file to the guardrail
addPIIFilter() Adds a PII filter to the guardrail
addRegexFilter() Adds a regex filter to the guardrail
addContextualGroundingFilter() Adds a contextual grounding filter to the guardrail
createVersion() Creates a new version of the guardrail
Guardrail Permissions

Guardrails provide methods to grant permissions to other resources that need to interact with the guardrail.

Permission Methods
Method Description Parameters
grant(grantee, ...actions) Grants the given principal identity permissions to perform actions on this guardrail grantee: The principal to grant permissions to
actions: The actions to grant (e.g., bedrock:GetGuardrail, bedrock:ListGuardrails)
grantApply(grantee) Grants the given identity permissions to apply the guardrail grantee: The principal to grant permissions to
Permission Examples
guardrail := bedrock.NewGuardrail(this, jsii.String("bedrockGuardrails"), &GuardrailProps{
	GuardrailName: jsii.String("my-BedrockGuardrails"),
})

lambdaFunction := lambda.NewFunction(this, jsii.String("testLambda"), &FunctionProps{
	Runtime: lambda.Runtime_PYTHON_3_12(),
	Handler: jsii.String("index.handler"),
	Code: lambda.Code_FromAsset(path.join(__dirname, jsii.String("../lambda/my-code"))),
})

// Grant specific permissions to a Lambda function
guardrail.Grant(lambdaFunction, jsii.String("bedrock:GetGuardrail"), jsii.String("bedrock:ListGuardrails"))

// Grant permissions to apply the guardrail
guardrail.GrantApply(lambdaFunction)
Guardrail Metrics

Amazon Bedrock provides metrics for your guardrails, allowing you to monitor their effectiveness and usage. These metrics are available in CloudWatch and can be used to create dashboards and alarms.

Metrics Examples
import cloudwatch "github.com/aws/aws-cdk-go/awscdk"


guardrail := bedrock.NewGuardrail(this, jsii.String("bedrockGuardrails"), &GuardrailProps{
	GuardrailName: jsii.String("my-BedrockGuardrails"),
})
// Get a specific metric for this guardrail
invocationsMetric := guardrail.MetricInvocations(&MetricOptions{
	Statistic: jsii.String("Sum"),
	Period: awscdk.Duration_Minutes(jsii.Number(5)),
})

// Create a CloudWatch alarm for high invocation latency
// Create a CloudWatch alarm for high invocation latency
cloudwatch.NewAlarm(this, jsii.String("HighLatencyAlarm"), &AlarmProps{
	Metric: guardrail.MetricInvocationLatency(),
	Threshold: jsii.Number(1000),
	 // 1 second
	EvaluationPeriods: jsii.Number(3),
})

// Get metrics for all guardrails
allInvocationsMetric := bedrock.Guardrail_MetricAllInvocations()
Importing Guardrails

You can import existing guardrails using the fromGuardrailAttributes or fromCfnGuardrail methods.

Import Configuration
var stack Stack

cmk := kms.NewKey(this, jsii.String("cmk"), &KeyProps{
})
// Import an existing guardrail by ARN
importedGuardrail := bedrock.Guardrail_FromGuardrailAttributes(stack, jsii.String("TestGuardrail"), &GuardrailAttributes{
	GuardrailArn: jsii.String("arn:aws:bedrock:us-east-1:123456789012:guardrail/oygh3o8g7rtl"),
	GuardrailVersion: jsii.String("1"),
	 //optional
	KmsKey: cmk,
})

import bedrockl1 "github.com/aws/aws-cdk-go/awscdk"

// Import a guardrail created through the L1 CDK CfnGuardrail construct
l1guardrail := bedrockl1.NewCfnGuardrail(this, jsii.String("MyCfnGuardrail"), &CfnGuardrailProps{
	BlockedInputMessaging: jsii.String("blockedInputMessaging"),
	BlockedOutputsMessaging: jsii.String("blockedOutputsMessaging"),
	Name: jsii.String("namemycfnguardrails"),
	WordPolicyConfig: &WordPolicyConfigProperty{
		WordsConfig: []interface{}{
			&WordConfigProperty{
				Text: jsii.String("drugs"),
			},
		},
	},
})

importedGuardrail := bedrock.Guardrail_FromCfnGuardrail(l1guardrail)
Guardrail Versioning

Guardrails support versioning, allowing you to track changes and maintain multiple versions of your guardrail configurations.

Version Configuration
guardrail := bedrock.NewGuardrail(this, jsii.String("bedrockGuardrails"), &GuardrailProps{
	GuardrailName: jsii.String("my-BedrockGuardrails"),
})
// Create a new version of the guardrail
guardrail.CreateVersion(jsii.String("testversion"))

Prompts

Amazon Bedrock provides the ability to create and save prompts using Prompt management so that you can save time by applying the same prompt to different workflows. You can include variables in the prompt so that you can adjust the prompt for different use case.

The Prompt resource allows you to create a new prompt.

Prompt Variants

Prompt variants in the context of Amazon Bedrock refer to alternative configurations of a prompt, including its message or the model and inference configurations used. Prompt variants are the building blocks of prompts - you must create at least one prompt variant to create a prompt. Prompt variants allow you to create different versions of a prompt, test them, and save the variant that works best for your use case.

There are three types of prompt variants:

  • Basic Text Prompt: Simple text-based prompts for straightforward interactions
  • Chat variant: Conversational prompts that support system messages, user/assistant message history, and tools
  • Agent variant: Prompts designed to work with Bedrock Agents
Basic Text Prompt

Text prompts are the simplest form of prompts, consisting of plain text instructions with optional variables. They are ideal for straightforward tasks like summarization, content generation, or question answering where you need a direct text-based interaction with the model.

cmk := kms.NewKey(this, jsii.String("cmk"), &KeyProps{
})
claudeModel := bedrock.BedrockFoundationModel_ANTHROPIC_CLAUDE_SONNET_V1_0()

variant1 := bedrock.PromptVariant_Text(&TextPromptVariantProps{
	VariantName: jsii.String("variant1"),
	Model: claudeModel,
	PromptVariables: []*string{
		jsii.String("topic"),
	},
	PromptText: jsii.String("This is my first text prompt. Please summarize our conversation on: {{topic}}."),
	InferenceConfiguration: bedrock.PromptInferenceConfiguration_Text(&PromptInferenceConfigurationProps{
		Temperature: jsii.Number(1),
		TopP: jsii.Number(0.999),
		MaxTokens: jsii.Number(2000),
	}),
})

prompt1 := bedrock.NewPrompt(this, jsii.String("prompt1"), &PromptProps{
	PromptName: jsii.String("prompt1"),
	Description: jsii.String("my first prompt"),
	DefaultVariant: variant1,
	Variants: []IPromptVariant{
		variant1,
	},
	KmsKey: cmk,
})
Chat Prompt

Use this template type when the model supports the Converse API or the Anthropic Claude Messages API. This allows you to include a System prompt and previous User messages and Assistant messages for context.

cmk := kms.NewKey(this, jsii.String("cmk"), &KeyProps{
})

variantChat := bedrock.PromptVariant_Chat(&ChatPromptVariantProps{
	VariantName: jsii.String("variant1"),
	Model: bedrock.BedrockFoundationModel_ANTHROPIC_CLAUDE_3_5_SONNET_V1_0(),
	Messages: []ChatMessage{
		bedrock.ChatMessage_User(jsii.String("From now on, you speak Japanese!")),
		bedrock.ChatMessage_Assistant(jsii.String("Konnichiwa!")),
		bedrock.ChatMessage_*User(jsii.String("From now on, you speak {{language}}!")),
	},
	System: jsii.String("You are a helpful assistant that only speaks the language you`re told."),
	PromptVariables: []*string{
		jsii.String("language"),
	},
	ToolConfiguration: &ToolConfiguration{
		ToolChoice: bedrock.ToolChoice_AUTO(),
		Tools: []Tool{
			bedrock.Tool_Function(&FunctionToolProps{
				Name: jsii.String("top_song"),
				Description: jsii.String("Get the most popular song played on a radio station."),
				InputSchema: map[string]interface{}{
					"type": jsii.String("object"),
					"properties": map[string]map[string]*string{
						"sign": map[string]*string{
							"type": jsii.String("string"),
							"description": jsii.String("The call sign for the radio station for which you want the most popular song. Example calls signs are WZPZ and WKR."),
						},
					},
					"required": []*string{
						jsii.String("sign"),
					},
				},
			}),
		},
	},
})

bedrock.NewPrompt(this, jsii.String("prompt1"), &PromptProps{
	PromptName: jsii.String("prompt-chat"),
	Description: jsii.String("my first chat prompt"),
	DefaultVariant: variantChat,
	Variants: []IPromptVariant{
		variantChat,
	},
	KmsKey: cmk,
})
Agent Prompt

Agent prompts are designed to work with Bedrock Agents, allowing you to create prompts that can be used by agents to perform specific tasks. Agent prompts use text prompts as their foundation and can reference agent aliases and include custom instructions for how the agent should behave.

cmk := kms.NewKey(this, jsii.String("cmk"), &KeyProps{
})

// Assuming you have an existing agent and alias
agent := bedrock.Agent_FromAgentAttributes(this, jsii.String("ImportedAgent"), &AgentAttributes{
	AgentArn: jsii.String("arn:aws:bedrock:region:account:agent/agent-id"),
	RoleArn: jsii.String("arn:aws:iam::account:role/agent-role"),
})

agentAlias := bedrock.AgentAlias_FromAttributes(this, jsii.String("ImportedAlias"), &AgentAliasAttributes{
	AliasId: jsii.String("alias-id"),
	AliasName: jsii.String("my-alias"),
	AgentVersion: jsii.String("1"),
	Agent: agent,
})

agentVariant := bedrock.PromptVariant_Agent(&AgentPromptVariantProps{
	VariantName: jsii.String("agent-variant"),
	Model: bedrock.BedrockFoundationModel_ANTHROPIC_CLAUDE_3_5_SONNET_V1_0(),
	AgentAlias: agentAlias,
	PromptText: jsii.String("Use the agent to help with: {{task}}. Please be thorough and provide detailed explanations."),
	PromptVariables: []*string{
		jsii.String("task"),
	},
})

bedrock.NewPrompt(this, jsii.String("agentPrompt"), &PromptProps{
	PromptName: jsii.String("agent-prompt"),
	Description: jsii.String("Prompt for agent interactions"),
	DefaultVariant: agentVariant,
	Variants: []IPromptVariant{
		agentVariant,
	},
	KmsKey: cmk,
})
Prompt Properties
Property Type Required Description
promptName string Yes The name of the prompt
description string No A description of the prompt
defaultVariant PromptVariant Yes The default variant to use for the prompt
variants PromptVariant[] No Additional variants for the prompt
kmsKey kms.IKey No The KMS key to use for encrypting the prompt. Defaults to AWS managed key
tags Record<string, string> No Tags to apply to the prompt
Prompt Version

A prompt version is a snapshot of a prompt at a specific point in time that you create when you are satisfied with a set of configurations. Versions allow you to deploy your prompt and easily switch between different configurations for your prompt and update your application with the most appropriate version for your use-case.

You can create a Prompt version by using the PromptVersion class or by using the .createVersion(..) on a Prompt object. It is recommended to use the .createVersion(..) method. It uses a hash based mechanism to update the version whenever a certain configuration property changes.

cmk := kms.NewKey(this, jsii.String("cmk"), &KeyProps{
})
claudeModel := bedrock.BedrockFoundationModel_ANTHROPIC_CLAUDE_SONNET_V1_0()

variant1 := bedrock.PromptVariant_Text(&TextPromptVariantProps{
	VariantName: jsii.String("variant1"),
	Model: claudeModel,
	PromptVariables: []*string{
		jsii.String("topic"),
	},
	PromptText: jsii.String("This is my first text prompt. Please summarize our conversation on: {{topic}}."),
	InferenceConfiguration: bedrock.PromptInferenceConfiguration_Text(&PromptInferenceConfigurationProps{
		Temperature: jsii.Number(1),
		TopP: jsii.Number(0.999),
		MaxTokens: jsii.Number(2000),
	}),
})

prompt1 := bedrock.NewPrompt(this, jsii.String("prompt1"), &PromptProps{
	PromptName: jsii.String("prompt1"),
	Description: jsii.String("my first prompt"),
	DefaultVariant: variant1,
	Variants: []IPromptVariant{
		variant1,
	},
	KmsKey: cmk,
})

promptVersion := bedrock.NewPromptVersion(this, jsii.String("MyPromptVersion"), &PromptVersionProps{
	Prompt: prompt1,
	Description: jsii.String("my first version"),
})
//or alternatively:
// const promptVersion = prompt1.createVersion('my first version');
versionString := promptVersion.Version
Import Methods

You can use the fromPromptAttributes method to import an existing Bedrock Prompt into your CDK application.

// Import an existing prompt by ARN
importedPrompt := bedrock.Prompt_FromPromptAttributes(this, jsii.String("ImportedPrompt"), &PromptAttributes{
	PromptArn: jsii.String("arn:aws:bedrock:region:account:prompt/prompt-id"),
	KmsKey: kms.Key_FromKeyArn(this, jsii.String("ImportedKey"), jsii.String("arn:aws:kms:region:account:key/key-id")),
	 // optional
	PromptVersion: jsii.String("1"),
})

Inference Profiles

Amazon Bedrock Inference Profiles provide a way to manage and optimize inference configurations for your foundation models. They allow you to define reusable configurations that can be applied across different prompts and agents.

Using Inference Profiles

Inference profiles can be used with prompts and agents to maintain consistent inference configurations across your application.

With Agents
// Create a cross-region inference profile
crossRegionProfile := bedrock.CrossRegionInferenceProfile_FromConfig(&CrossRegionInferenceProfileProps{
	GeoRegion: bedrock.CrossRegionInferenceProfileRegion_US,
	Model: bedrock.BedrockFoundationModel_ANTHROPIC_CLAUDE_3_5_SONNET_V1_0(),
})

// Use the cross-region profile with an agent
agent := bedrock.NewAgent(this, jsii.String("Agent"), &AgentProps{
	FoundationModel: crossRegionProfile,
	Instruction: jsii.String("You are a helpful and friendly agent that answers questions about agriculture."),
})
With Prompts
// Create a prompt router for intelligent model selection
promptRouter := bedrock.PromptRouter_FromDefaultId(bedrock.DefaultPromptRouterIdentifier_ANTHROPIC_CLAUDE_V1(), jsii.String("us-east-1"))

// Use the prompt router with a prompt variant
variant := bedrock.PromptVariant_Text(&TextPromptVariantProps{
	VariantName: jsii.String("variant1"),
	PromptText: jsii.String("What is the capital of France?"),
	Model: promptRouter,
})

bedrock.NewPrompt(this, jsii.String("Prompt"), &PromptProps{
	PromptName: jsii.String("prompt-router-test"),
	Variants: []IPromptVariant{
		variant,
	},
})
Types of Inference Profiles

Amazon Bedrock offers two types of inference profiles:

Application Inference Profiles

Application inference profiles are user-defined profiles that help you track costs and model usage. They can be created for a single region or for multiple regions using a cross-region inference profile.

Single Region Application Profile
// Create an application inference profile for one Region
appProfile := bedrock.NewApplicationInferenceProfile(this, jsii.String("MyApplicationProfile"), &ApplicationInferenceProfileProps{
	ApplicationInferenceProfileName: jsii.String("claude-3-sonnet-v1"),
	ModelSource: bedrock.BedrockFoundationModel_ANTHROPIC_CLAUDE_SONNET_V1_0(),
	Description: jsii.String("Application profile for cost tracking"),
	Tags: map[string]*string{
		"Environment": jsii.String("Production"),
	},
})
Multi-Region Application Profile
// Create a cross-region inference profile
crossRegionProfile := bedrock.CrossRegionInferenceProfile_FromConfig(&CrossRegionInferenceProfileProps{
	GeoRegion: bedrock.CrossRegionInferenceProfileRegion_US,
	Model: bedrock.BedrockFoundationModel_ANTHROPIC_CLAUDE_3_5_SONNET_V2_0(),
})

// Create an application inference profile across regions
appProfile := bedrock.NewApplicationInferenceProfile(this, jsii.String("MyMultiRegionProfile"), &ApplicationInferenceProfileProps{
	ApplicationInferenceProfileName: jsii.String("claude-35-sonnet-v2-multi-region"),
	ModelSource: crossRegionProfile,
	Description: jsii.String("Multi-region application profile for cost tracking"),
})
System Defined Inference Profiles

Cross-region inference enables you to seamlessly manage unplanned traffic bursts by utilizing compute across different AWS Regions. With cross-region inference, you can distribute traffic across multiple AWS Regions, enabling higher throughput and enhanced resilience during periods of peak demands.

Before using a CrossRegionInferenceProfile, ensure that you have access to the models and regions defined in the inference profiles. For instance, if you use the system defined inference profile "us.anthropic.claude-3-5-sonnet-20241022-v2:0", inference requests will be routed to US East (Virginia) us-east-1, US East (Ohio) us-east-2 and US West (Oregon) us-west-2. Thus, you need to have model access enabled in those regions for the model anthropic.claude-3-5-sonnet-20241022-v2:0.

System Defined Profile Configuration
crossRegionProfile := bedrock.CrossRegionInferenceProfile_FromConfig(&CrossRegionInferenceProfileProps{
	GeoRegion: bedrock.CrossRegionInferenceProfileRegion_US,
	Model: bedrock.BedrockFoundationModel_ANTHROPIC_CLAUDE_3_5_SONNET_V2_0(),
})
Prompt Routers

Amazon Bedrock intelligent prompt routing provides a single serverless endpoint for efficiently routing requests between different foundational models within the same model family. It can help you optimize for response quality and cost. They offer a comprehensive solution for managing multiple AI models through a single serverless endpoint, simplifying the process for you. Intelligent prompt routing predicts the performance of each model for each request, and dynamically routes each request to the model that it predicts is most likely to give the desired response at the lowest cost.

Default and Custom Prompt Routers
// Use a default prompt router
variant := bedrock.PromptVariant_Text(&TextPromptVariantProps{
	VariantName: jsii.String("variant1"),
	PromptText: jsii.String("What is the capital of France?"),
	Model: bedrock.PromptRouter_FromDefaultId(bedrock.DefaultPromptRouterIdentifier_ANTHROPIC_CLAUDE_V1(), jsii.String("us-east-1")),
})

bedrock.NewPrompt(this, jsii.String("Prompt"), &PromptProps{
	PromptName: jsii.String("prompt-router-test"),
	Variants: []IPromptVariant{
		variant,
	},
})
Inference Profile Permissions

Use the grantProfileUsage method to grant appropriate permissions to resources that need to use the inference profile.

Granting Profile Usage Permissions
// Create an application inference profile
profile := bedrock.NewApplicationInferenceProfile(this, jsii.String("MyProfile"), &ApplicationInferenceProfileProps{
	ApplicationInferenceProfileName: jsii.String("my-profile"),
	ModelSource: bedrock.BedrockFoundationModel_ANTHROPIC_CLAUDE_SONNET_4_5_V1_0(),
})

// Create a Lambda function
lambdaFunction := lambda.NewFunction(this, jsii.String("MyFunction"), &FunctionProps{
	Runtime: lambda.Runtime_PYTHON_3_11(),
	Handler: jsii.String("index.handler"),
	Code: lambda.Code_FromInline(jsii.String("def handler(event, context): return \"Hello\"")),
})

// Grant the Lambda function permission to use the inference profile
profile.GrantProfileUsage(lambdaFunction)

// Use a system defined inference profile
crossRegionProfile := bedrock.CrossRegionInferenceProfile_FromConfig(&CrossRegionInferenceProfileProps{
	GeoRegion: bedrock.CrossRegionInferenceProfileRegion_US,
	Model: bedrock.BedrockFoundationModel_ANTHROPIC_CLAUDE_SONNET_4_5_V1_0(),
})

// Grant permissions to use the cross-region inference profile
crossRegionProfile.GrantProfileUsage(lambdaFunction)

The grantProfileUsage method adds the necessary IAM permissions to the resource, allowing it to use the inference profile. This includes permissions to call bedrock:GetInferenceProfile and bedrock:ListInferenceProfiles actions on the inference profile resource.

Inference Profiles Import Methods

You can import existing application inference profiles using the following methods:

// Import an inference profile through attributes
importedProfile := bedrock.ApplicationInferenceProfile_FromApplicationInferenceProfileAttributes(this, jsii.String("ImportedProfile"), &ApplicationInferenceProfileAttributes{
	InferenceProfileArn: jsii.String("arn:aws:bedrock:us-east-1:123456789012:application-inference-profile/my-profile-id"),
	InferenceProfileIdentifier: jsii.String("my-profile-id"),
})

You can also import an application inference profile from an existing L1 CloudFormation construct:

// Create or reference an existing L1 CfnApplicationInferenceProfile
cfnProfile := awscdk.Aws_bedrock.NewCfnApplicationInferenceProfile(this, jsii.String("CfnProfile"), &CfnApplicationInferenceProfileProps{
	InferenceProfileName: jsii.String("my-cfn-profile"),
	ModelSource: &InferenceProfileModelSourceProperty{
		CopyFrom: bedrock.BedrockFoundationModel_ANTHROPIC_CLAUDE_3_5_SONNET_V1_0().InvokableArn,
	},
	Description: jsii.String("Profile created via L1 construct"),
})

// Import the L1 construct as an L2 ApplicationInferenceProfile
importedFromCfn := bedrock.ApplicationInferenceProfile_FromCfnApplicationInferenceProfile(cfnProfile)

// Grant permissions to use the imported profile
lambdaFunction := lambda.NewFunction(this, jsii.String("MyFunction"), &FunctionProps{
	Runtime: lambda.Runtime_PYTHON_3_11(),
	Handler: jsii.String("index.handler"),
	Code: lambda.Code_FromInline(jsii.String("def handler(event, context): return \"Hello\"")),
})

importedFromCfn.GrantProfileUsage(lambdaFunction)

Documentation

Overview

The CDK Construct Library for Amazon Bedrock

Index

Constants

This section is empty.

Variables

This section is empty.

Functions

func AgentAliasBase_IsConstruct 

func AgentAliasBase_IsConstruct(x interface{}) *bool

Checks if `x` is a construct.

Use this method instead of `instanceof` to properly detect `Construct` instances, even when the construct library is symlinked.

Explanation: in JavaScript, multiple copies of the `constructs` library on disk are seen as independent, completely different libraries. As a consequence, the class `Construct` in each copy of the `constructs` library is seen as a different class, and an instance of one class will not test as `instanceof` the other class. `npm install` will not create installations like this, but users may manually symlink construct libraries together or use a monorepo tool: in those cases, multiple copies of the `constructs` library can be accidentally installed, and `instanceof` will behave unpredictably. It is safest to avoid using `instanceof`, and using this type-testing method instead.

Returns: true if `x` is an object created from a class which extends `Construct`. Experimental.

func AgentAliasBase_IsOwnedResource 

func AgentAliasBase_IsOwnedResource(construct constructs.IConstruct) *bool

Returns true if the construct was created by CDK, and false otherwise. Experimental.

func AgentAliasBase_IsResource 

func AgentAliasBase_IsResource(construct constructs.IConstruct) *bool

Check whether the given construct is a Resource. Experimental.

func AgentAlias_IsConstruct 

func AgentAlias_IsConstruct(x interface{}) *bool

Checks if `x` is a construct.

Use this method instead of `instanceof` to properly detect `Construct` instances, even when the construct library is symlinked.

Explanation: in JavaScript, multiple copies of the `constructs` library on disk are seen as independent, completely different libraries. As a consequence, the class `Construct` in each copy of the `constructs` library is seen as a different class, and an instance of one class will not test as `instanceof` the other class. `npm install` will not create installations like this, but users may manually symlink construct libraries together or use a monorepo tool: in those cases, multiple copies of the `constructs` library can be accidentally installed, and `instanceof` will behave unpredictably. It is safest to avoid using `instanceof`, and using this type-testing method instead.

Returns: true if `x` is an object created from a class which extends `Construct`. Experimental.

func AgentAlias_IsOwnedResource 

func AgentAlias_IsOwnedResource(construct constructs.IConstruct) *bool

Returns true if the construct was created by CDK, and false otherwise. Experimental.

func AgentAlias_IsResource 

func AgentAlias_IsResource(construct constructs.IConstruct) *bool

Check whether the given construct is a Resource. Experimental.

func AgentAlias_PROPERTY_INJECTION_ID 

func AgentAlias_PROPERTY_INJECTION_ID() *string

func AgentBase_IsConstruct 

func AgentBase_IsConstruct(x interface{}) *bool

Checks if `x` is a construct.

Use this method instead of `instanceof` to properly detect `Construct` instances, even when the construct library is symlinked.

Explanation: in JavaScript, multiple copies of the `constructs` library on disk are seen as independent, completely different libraries. As a consequence, the class `Construct` in each copy of the `constructs` library is seen as a different class, and an instance of one class will not test as `instanceof` the other class. `npm install` will not create installations like this, but users may manually symlink construct libraries together or use a monorepo tool: in those cases, multiple copies of the `constructs` library can be accidentally installed, and `instanceof` will behave unpredictably. It is safest to avoid using `instanceof`, and using this type-testing method instead.

Returns: true if `x` is an object created from a class which extends `Construct`. Experimental.

func AgentBase_IsOwnedResource 

func AgentBase_IsOwnedResource(construct constructs.IConstruct) *bool

Returns true if the construct was created by CDK, and false otherwise. Experimental.

func AgentBase_IsResource 

func AgentBase_IsResource(construct constructs.IConstruct) *bool

Check whether the given construct is a Resource. Experimental.

func Agent_IsConstruct 

func Agent_IsConstruct(x interface{}) *bool

Checks if `x` is a construct.

Use this method instead of `instanceof` to properly detect `Construct` instances, even when the construct library is symlinked.

Explanation: in JavaScript, multiple copies of the `constructs` library on disk are seen as independent, completely different libraries. As a consequence, the class `Construct` in each copy of the `constructs` library is seen as a different class, and an instance of one class will not test as `instanceof` the other class. `npm install` will not create installations like this, but users may manually symlink construct libraries together or use a monorepo tool: in those cases, multiple copies of the `constructs` library can be accidentally installed, and `instanceof` will behave unpredictably. It is safest to avoid using `instanceof`, and using this type-testing method instead.

Returns: true if `x` is an object created from a class which extends `Construct`. Experimental.

func Agent_IsOwnedResource 

func Agent_IsOwnedResource(construct constructs.IConstruct) *bool

Returns true if the construct was created by CDK, and false otherwise. Experimental.

func Agent_IsResource 

func Agent_IsResource(construct constructs.IConstruct) *bool

Check whether the given construct is a Resource. Experimental.

func Agent_PROPERTY_INJECTION_ID 

func Agent_PROPERTY_INJECTION_ID() *string

func ApplicationInferenceProfile_IsConstruct 

func ApplicationInferenceProfile_IsConstruct(x interface{}) *bool

Checks if `x` is a construct.

Use this method instead of `instanceof` to properly detect `Construct` instances, even when the construct library is symlinked.

Explanation: in JavaScript, multiple copies of the `constructs` library on disk are seen as independent, completely different libraries. As a consequence, the class `Construct` in each copy of the `constructs` library is seen as a different class, and an instance of one class will not test as `instanceof` the other class. `npm install` will not create installations like this, but users may manually symlink construct libraries together or use a monorepo tool: in those cases, multiple copies of the `constructs` library can be accidentally installed, and `instanceof` will behave unpredictably. It is safest to avoid using `instanceof`, and using this type-testing method instead.

Returns: true if `x` is an object created from a class which extends `Construct`. Experimental.

func ApplicationInferenceProfile_IsOwnedResource 

func ApplicationInferenceProfile_IsOwnedResource(construct constructs.IConstruct) *bool

Returns true if the construct was created by CDK, and false otherwise. Experimental.

func ApplicationInferenceProfile_IsResource 

func ApplicationInferenceProfile_IsResource(construct constructs.IConstruct) *bool

Check whether the given construct is a Resource. Experimental.

func ApplicationInferenceProfile_PROPERTY_INJECTION_ID 

func ApplicationInferenceProfile_PROPERTY_INJECTION_ID() *string

func GuardrailBase_IsConstruct 

func GuardrailBase_IsConstruct(x interface{}) *bool

Checks if `x` is a construct.

Use this method instead of `instanceof` to properly detect `Construct` instances, even when the construct library is symlinked.

Explanation: in JavaScript, multiple copies of the `constructs` library on disk are seen as independent, completely different libraries. As a consequence, the class `Construct` in each copy of the `constructs` library is seen as a different class, and an instance of one class will not test as `instanceof` the other class. `npm install` will not create installations like this, but users may manually symlink construct libraries together or use a monorepo tool: in those cases, multiple copies of the `constructs` library can be accidentally installed, and `instanceof` will behave unpredictably. It is safest to avoid using `instanceof`, and using this type-testing method instead.

Returns: true if `x` is an object created from a class which extends `Construct`. Experimental.

func GuardrailBase_IsOwnedResource 

func GuardrailBase_IsOwnedResource(construct constructs.IConstruct) *bool

Returns true if the construct was created by CDK, and false otherwise. Experimental.

func GuardrailBase_IsResource 

func GuardrailBase_IsResource(construct constructs.IConstruct) *bool

Check whether the given construct is a Resource. Experimental.

func GuardrailBase_MetricAll 

func GuardrailBase_MetricAll(metricName *string, props *awscloudwatch.MetricOptions) awscloudwatch.Metric

Return the given named metric for all guardrails.

By default, the metric will be calculated as a sum over a period of 5 minutes. You can customize this by using the `statistic` and `period` properties. Experimental.

func GuardrailBase_MetricAllInvocationLatency 

func GuardrailBase_MetricAllInvocationLatency(props *awscloudwatch.MetricOptions) awscloudwatch.Metric

Return the invocation latency metric for all guardrails. Experimental.

func GuardrailBase_MetricAllInvocations 

func GuardrailBase_MetricAllInvocations(props *awscloudwatch.MetricOptions) awscloudwatch.Metric

Return the invocations metric for all guardrails. Experimental.

func GuardrailBase_MetricAllInvocationsIntervened 

func GuardrailBase_MetricAllInvocationsIntervened(props *awscloudwatch.MetricOptions) awscloudwatch.Metric

Return the invocations intervened metric for all guardrails. Experimental.

func GuardrailBase_MetricAllTextUnitCount 

func GuardrailBase_MetricAllTextUnitCount(props *awscloudwatch.MetricOptions) awscloudwatch.Metric

Return the text unit count metric for all guardrails. Experimental.

func Guardrail_IsConstruct 

func Guardrail_IsConstruct(x interface{}) *bool

Checks if `x` is a construct.

Use this method instead of `instanceof` to properly detect `Construct` instances, even when the construct library is symlinked.

Explanation: in JavaScript, multiple copies of the `constructs` library on disk are seen as independent, completely different libraries. As a consequence, the class `Construct` in each copy of the `constructs` library is seen as a different class, and an instance of one class will not test as `instanceof` the other class. `npm install` will not create installations like this, but users may manually symlink construct libraries together or use a monorepo tool: in those cases, multiple copies of the `constructs` library can be accidentally installed, and `instanceof` will behave unpredictably. It is safest to avoid using `instanceof`, and using this type-testing method instead.

Returns: true if `x` is an object created from a class which extends `Construct`. Experimental.

func Guardrail_IsOwnedResource 

func Guardrail_IsOwnedResource(construct constructs.IConstruct) *bool

Returns true if the construct was created by CDK, and false otherwise. Experimental.

func Guardrail_IsResource 

func Guardrail_IsResource(construct constructs.IConstruct) *bool

Check whether the given construct is a Resource. Experimental.

func Guardrail_MetricAll 

func Guardrail_MetricAll(metricName *string, props *awscloudwatch.MetricOptions) awscloudwatch.Metric

Return the given named metric for all guardrails.

By default, the metric will be calculated as a sum over a period of 5 minutes. You can customize this by using the `statistic` and `period` properties. Experimental.

func Guardrail_MetricAllInvocationLatency 

func Guardrail_MetricAllInvocationLatency(props *awscloudwatch.MetricOptions) awscloudwatch.Metric

Return the invocation latency metric for all guardrails. Experimental.

func Guardrail_MetricAllInvocations 

func Guardrail_MetricAllInvocations(props *awscloudwatch.MetricOptions) awscloudwatch.Metric

Return the invocations metric for all guardrails. Experimental.

func Guardrail_MetricAllInvocationsIntervened 

func Guardrail_MetricAllInvocationsIntervened(props *awscloudwatch.MetricOptions) awscloudwatch.Metric

Return the invocations intervened metric for all guardrails. Experimental.

func Guardrail_MetricAllTextUnitCount 

func Guardrail_MetricAllTextUnitCount(props *awscloudwatch.MetricOptions) awscloudwatch.Metric

Return the text unit count metric for all guardrails. Experimental.

func Guardrail_PROPERTY_INJECTION_ID 

func Guardrail_PROPERTY_INJECTION_ID() *string

func InferenceProfileBase_IsConstruct 

func InferenceProfileBase_IsConstruct(x interface{}) *bool

Checks if `x` is a construct.

Use this method instead of `instanceof` to properly detect `Construct` instances, even when the construct library is symlinked.

Explanation: in JavaScript, multiple copies of the `constructs` library on disk are seen as independent, completely different libraries. As a consequence, the class `Construct` in each copy of the `constructs` library is seen as a different class, and an instance of one class will not test as `instanceof` the other class. `npm install` will not create installations like this, but users may manually symlink construct libraries together or use a monorepo tool: in those cases, multiple copies of the `constructs` library can be accidentally installed, and `instanceof` will behave unpredictably. It is safest to avoid using `instanceof`, and using this type-testing method instead.

Returns: true if `x` is an object created from a class which extends `Construct`. Experimental.

func InferenceProfileBase_IsOwnedResource 

func InferenceProfileBase_IsOwnedResource(construct constructs.IConstruct) *bool

Returns true if the construct was created by CDK, and false otherwise. Experimental.

func InferenceProfileBase_IsResource 

func InferenceProfileBase_IsResource(construct constructs.IConstruct) *bool

Check whether the given construct is a Resource. Experimental.

func NewAgentActionGroup_Override 

func NewAgentActionGroup_Override(a AgentActionGroup, props *AgentActionGroupProps)

Experimental.

func NewAgentAliasBase_Override 

func NewAgentAliasBase_Override(a AgentAliasBase, scope constructs.Construct, id *string, props *awscdk.ResourceProps)

Experimental.

func NewAgentAlias_Override 

func NewAgentAlias_Override(a AgentAlias, scope constructs.Construct, id *string, props *AgentAliasProps)

Experimental.

func NewAgentBase_Override 

func NewAgentBase_Override(a AgentBase, scope constructs.Construct, id *string, props *awscdk.ResourceProps)

Experimental.

func NewAgentCollaboration_Override 

func NewAgentCollaboration_Override(a AgentCollaboration, config *AgentCollaborationConfig)

Experimental.

func NewAgentCollaborator_Override 

func NewAgentCollaborator_Override(a AgentCollaborator, props *AgentCollaboratorProps)

Experimental.

func NewAgent_Override 

func NewAgent_Override(a Agent, scope constructs.Construct, id *string, props *AgentProps)

Experimental.

func NewApiSchema_Override 

func NewApiSchema_Override(a ApiSchema, s3File *awss3.Location, inlineSchema *string)

Experimental.

func NewApplicationInferenceProfile_Override 

func NewApplicationInferenceProfile_Override(a ApplicationInferenceProfile, scope constructs.Construct, id *string, props *ApplicationInferenceProfileProps)

Experimental.

func NewAssetApiSchema_Override 

func NewAssetApiSchema_Override(a AssetApiSchema, path *string, options *awss3assets.AssetOptions)

Experimental.

func NewBedrockFoundationModel_Override 

func NewBedrockFoundationModel_Override(b BedrockFoundationModel, value *string, props *BedrockFoundationModelProps)

Experimental.

func NewChatMessage_Override 

func NewChatMessage_Override(c ChatMessage, role ChatMessageRole, text *string)

Experimental.

func NewFunctionParameter_Override 

func NewFunctionParameter_Override(f FunctionParameter, props *FunctionParameterProps)

Experimental.

func NewFunctionSchema_Override 

func NewFunctionSchema_Override(f FunctionSchema, props *FunctionSchemaProps)

Experimental.

func NewFunction_Override 

func NewFunction_Override(f Function, props *FunctionProps)

Experimental.

func NewGuardrailBase_Override 

func NewGuardrailBase_Override(g GuardrailBase, scope constructs.Construct, id *string, props *awscdk.ResourceProps)

Experimental.

func NewGuardrail_Override 

func NewGuardrail_Override(g Guardrail, scope constructs.Construct, id *string, props *GuardrailProps)

Experimental.

func NewInferenceProfileBase_Override 

func NewInferenceProfileBase_Override(i InferenceProfileBase, scope constructs.Construct, id *string, props *awscdk.ResourceProps)

Experimental.

func NewInlineApiSchema_Override 

func NewInlineApiSchema_Override(i InlineApiSchema, schema *string)

Experimental.

func NewMemory_Override 

func NewMemory_Override(m Memory, props *SessionSummaryMemoryProps)

Experimental.

func NewPIIType_Override 

func NewPIIType_Override(p PIIType, value *string)

The string value of the PII type. Experimental.

func NewParentActionGroupSignature_Override 

func NewParentActionGroupSignature_Override(p ParentActionGroupSignature, value *string)

Constructor should be used as a temporary solution when a new signature is supported but its implementation in CDK hasn't been added yet. Experimental.

func NewPromptBase_Override 

func NewPromptBase_Override(p PromptBase, scope constructs.Construct, id *string, props *awscdk.ResourceProps)

Experimental.

func NewPromptGenAiResource_Override 

func NewPromptGenAiResource_Override(p PromptGenAiResource)

Experimental.

func NewPromptInferenceConfiguration_Override 

func NewPromptInferenceConfiguration_Override(p PromptInferenceConfiguration)

Experimental.

func NewPromptRouter_Override 

func NewPromptRouter_Override(p PromptRouter, props *PromptRouterProps, region *string)

Experimental.

func NewPromptTemplateConfiguration_Override 

func NewPromptTemplateConfiguration_Override(p PromptTemplateConfiguration)

Experimental.

func NewPromptVersion_Override 

func NewPromptVersion_Override(p PromptVersion, scope constructs.Construct, id *string, props *PromptVersionProps)

**************************************************************************** CONSTRUCTOR ***************************************************************************. Experimental.

func NewPrompt_Override 

func NewPrompt_Override(p Prompt, scope constructs.Construct, id *string, props *PromptProps)

**************************************************************************** CONSTRUCTOR ***************************************************************************. Experimental.

func NewS3ApiSchema_Override 

func NewS3ApiSchema_Override(s S3ApiSchema, location *awss3.Location)

Experimental.

func NewToolChoice_Override 

func NewToolChoice_Override(t ToolChoice, any interface{}, auto interface{}, tool *string)

Experimental.

func NewTool_Override 

func NewTool_Override(t Tool)

Experimental.

func NewTopic_Override 

func NewTopic_Override(t Topic, props *CustomTopicProps)

Experimental.

func PromptBase_IsConstruct 

func PromptBase_IsConstruct(x interface{}) *bool

Checks if `x` is a construct.

Use this method instead of `instanceof` to properly detect `Construct` instances, even when the construct library is symlinked.

Explanation: in JavaScript, multiple copies of the `constructs` library on disk are seen as independent, completely different libraries. As a consequence, the class `Construct` in each copy of the `constructs` library is seen as a different class, and an instance of one class will not test as `instanceof` the other class. `npm install` will not create installations like this, but users may manually symlink construct libraries together or use a monorepo tool: in those cases, multiple copies of the `constructs` library can be accidentally installed, and `instanceof` will behave unpredictably. It is safest to avoid using `instanceof`, and using this type-testing method instead.

Returns: true if `x` is an object created from a class which extends `Construct`. Experimental.

func PromptBase_IsOwnedResource 

func PromptBase_IsOwnedResource(construct constructs.IConstruct) *bool

Returns true if the construct was created by CDK, and false otherwise. Experimental.

func PromptBase_IsResource 

func PromptBase_IsResource(construct constructs.IConstruct) *bool

Check whether the given construct is a Resource. Experimental.

func PromptVersion_IsConstruct 

func PromptVersion_IsConstruct(x interface{}) *bool

Checks if `x` is a construct.

Use this method instead of `instanceof` to properly detect `Construct` instances, even when the construct library is symlinked.

Explanation: in JavaScript, multiple copies of the `constructs` library on disk are seen as independent, completely different libraries. As a consequence, the class `Construct` in each copy of the `constructs` library is seen as a different class, and an instance of one class will not test as `instanceof` the other class. `npm install` will not create installations like this, but users may manually symlink construct libraries together or use a monorepo tool: in those cases, multiple copies of the `constructs` library can be accidentally installed, and `instanceof` will behave unpredictably. It is safest to avoid using `instanceof`, and using this type-testing method instead.

Returns: true if `x` is an object created from a class which extends `Construct`. Experimental.

func Prompt_IsConstruct 

func Prompt_IsConstruct(x interface{}) *bool

Checks if `x` is a construct.

Use this method instead of `instanceof` to properly detect `Construct` instances, even when the construct library is symlinked.

Explanation: in JavaScript, multiple copies of the `constructs` library on disk are seen as independent, completely different libraries. As a consequence, the class `Construct` in each copy of the `constructs` library is seen as a different class, and an instance of one class will not test as `instanceof` the other class. `npm install` will not create installations like this, but users may manually symlink construct libraries together or use a monorepo tool: in those cases, multiple copies of the `constructs` library can be accidentally installed, and `instanceof` will behave unpredictably. It is safest to avoid using `instanceof`, and using this type-testing method instead.

Returns: true if `x` is an object created from a class which extends `Construct`. Experimental.

func Prompt_IsOwnedResource 

func Prompt_IsOwnedResource(construct constructs.IConstruct) *bool

Returns true if the construct was created by CDK, and false otherwise. Experimental.

func Prompt_IsResource 

func Prompt_IsResource(construct constructs.IConstruct) *bool

Check whether the given construct is a Resource. Experimental.

func Prompt_PROPERTY_INJECTION_ID 

func Prompt_PROPERTY_INJECTION_ID() *string

Types

type ActionGroupExecutor 

type ActionGroupExecutor interface {
	// The custom control type for the action group executor.
	//
	// Currently only supports 'RETURN_CONTROL' which returns results directly in the InvokeAgent response.
	// Experimental.
	CustomControl() CustomControl
	// The Lambda function that will be called by the action group.
	//
	// Contains the business logic for handling the action group's invocation.
	// Experimental.
	LambdaFunction() awslambda.IFunction
}

Defines how fulfillment of the action group is handled after the necessary information has been elicited from the user.

Valid executors are: - Lambda function - Return Control.

Example: 

bucket := s3.Bucket_FromBucketName(this, jsii.String("ExistingBucket"), jsii.String("my-schema-bucket"))
s3Schema := bedrock.ApiSchema_FromS3File(bucket, jsii.String("schemas/action-group.yaml"))

actionGroupFunction := lambda.NewFunction(this, jsii.String("ActionGroupFunction"), &FunctionProps{
	Runtime: lambda.Runtime_PYTHON_3_12(),
	Handler: jsii.String("index.handler"),
	Code: lambda.Code_FromAsset(path.join(__dirname, jsii.String("../lambda/action-group"))),
})

actionGroup := bedrock.NewAgentActionGroup(&AgentActionGroupProps{
	Name: jsii.String("query-library"),
	Description: jsii.String("Use these functions to get information about the books in the library."),
	Executor: bedrock.ActionGroupExecutor_FromLambda(actionGroupFunction),
	Enabled: jsii.Boolean(true),
	ApiSchema: s3Schema,
})

agent := bedrock.NewAgent(this, jsii.String("Agent"), &AgentProps{
	FoundationModel: bedrock.BedrockFoundationModel_ANTHROPIC_CLAUDE_HAIKU_V1_0(),
	Instruction: jsii.String("You are a helpful and friendly agent that answers questions about literature."),
})

agent.AddActionGroup(actionGroup)

See: https://docs.aws.amazon.com/bedrock/latest/userguide/action-handle.html

Experimental.

func ActionGroupExecutor_FromLambda 

func ActionGroupExecutor_FromLambda(lambdaFunction awslambda.IFunction) ActionGroupExecutor

Defines an action group with a Lambda function containing the business logic. See: https://docs.aws.amazon.com/bedrock/latest/userguide/agents-lambda.html

Experimental.

func ActionGroupExecutor_RETURN_CONTROL 

func ActionGroupExecutor_RETURN_CONTROL() ActionGroupExecutor

type Agent 

type Agent interface {
	AgentBase
	IAgent
	// action groups associated with the ageny.
	// Experimental.
	ActionGroups() *[]AgentActionGroup
	// The ARN of the agent.
	// Experimental.
	AgentArn() *string
	// The unique identifier for the agent.
	// Experimental.
	AgentId() *string
	// The version of the agent.
	// Experimental.
	AgentVersion() *string
	// The environment this resource belongs to.
	//
	// For resources that are created and managed in a Stack (those created by
	// creating new class instances like `new Role()`, `new Bucket()`, etc.), this
	// is always the same as the environment of the stack they belong to.
	//
	// For referenced resources (those obtained from referencing methods like
	// `Role.fromRoleArn()`, `Bucket.fromBucketName()`, etc.), they might be
	// different than the stack they were imported into.
	// Experimental.
	Env() *interfaces.ResourceEnvironment
	// The principal to grant permissions to.
	// Experimental.
	GrantPrincipal() awsiam.IPrincipal
	// The guardrail that will be associated with the agent.
	// Experimental.
	Guardrail() IGuardrail
	// Experimental.
	SetGuardrail(val IGuardrail)
	// Optional KMS encryption key associated with this agent.
	// Experimental.
	KmsKey() awskms.IKey
	// When this agent was last updated.
	// Experimental.
	LastUpdated() *string
	// The name of the agent.
	// Experimental.
	Name() *string
	// The tree node.
	// Experimental.
	Node() constructs.Node
	// Returns a string-encoded token that resolves to the physical name that should be passed to the CloudFormation resource.
	//
	// This value will resolve to one of the following:
	// - a concrete value (e.g. `"my-awesome-bucket"`)
	// - `undefined`, when a name should be generated by CloudFormation
	// - a concrete name generated automatically during synthesis, in
	//   cross-environment scenarios.
	// Experimental.
	PhysicalName() *string
	// The IAM role associated to the agent.
	// Experimental.
	Role() awsiam.IRole
	// The stack in which this resource is defined.
	// Experimental.
	Stack() awscdk.Stack
	// Default alias of the agent.
	// Experimental.
	TestAlias() IAgentAlias
	// Adds an action group to the agent and configures necessary permissions.
	// Default: - Default permissions:
	// - Lambda function invoke permissions if executor is present
	// - S3 GetObject permissions if apiSchema.s3File is present
	//
	// Experimental.
	AddActionGroup(actionGroup AgentActionGroup)
	// Configuration for agent collaboration.
	// Default: - No collaboration configuration.
	//
	// Experimental.
	AddActionGroups(actionGroups ...AgentActionGroup)
	// Add guardrail to the agent.
	// Experimental.
	AddGuardrail(guardrail IGuardrail)
	// Apply the given removal policy to this resource.
	//
	// The Removal Policy controls what happens to this resource when it stops
	// being managed by CloudFormation, either because you've removed it from the
	// CDK application or because you've made a change that requires the resource
	// to be replaced.
	//
	// The resource can be deleted (`RemovalPolicy.DESTROY`), or left in your AWS
	// account for data recovery and cleanup later (`RemovalPolicy.RETAIN`).
	// Experimental.
	ApplyRemovalPolicy(policy awscdk.RemovalPolicy)
	// Generates a physical name for the agent.
	//
	// Returns: A unique name for the agent with appropriate length constraints.
	// Default: - Generated name format: 'agent-{hash}-{uniqueName}' with:
	// - maxLength: MAXLENGTH_FOR_ROLE_NAME - '-bedrockagent'.length
	// - lower: true
	// - separator: '-'.
	//
	// Experimental.
	GeneratePhysicalName() *string
	// Returns an environment-sensitive token that should be used for the resource's "ARN" attribute (e.g. `bucket.bucketArn`).
	//
	// Normally, this token will resolve to `arnAttr`, but if the resource is
	// referenced across environments, `arnComponents` will be used to synthesize
	// a concrete ARN with the resource's physical name. Make sure to reference
	// `this.physicalName` in `arnComponents`.
	// Experimental.
	GetResourceArnAttribute(arnAttr *string, arnComponents *awscdk.ArnComponents) *string
	// Returns an environment-sensitive token that should be used for the resource's "name" attribute (e.g. `bucket.bucketName`).
	//
	// Normally, this token will resolve to `nameAttr`, but if the resource is
	// referenced across environments, it will be resolved to `this.physicalName`,
	// which will be a concrete name.
	// Experimental.
	GetResourceNameAttribute(nameAttr *string) *string
	// Grant invoke permissions on this agent to an IAM principal.
	//
	// [disable-awslint:no-grants].
	//
	// Returns: An IAM Grant object representing the granted permissions.
	// Default: - Default grant configuration:
	// - actions: ['bedrock:InvokeAgent']
	// - resourceArns: [this.agentArn]
	//
	// Experimental.
	GrantInvoke(grantee awsiam.IGrantable) awsiam.Grant
	// Creates a CloudWatch metric for tracking agent invocations.
	//
	// Returns: A CloudWatch Metric configured for agent invocation counts.
	// Default: - Default metric configuration:
	// - namespace: 'AWS/Bedrock'
	// - metricName: 'Invocations'
	// - dimensionsMap: { AgentId: this.agentId }
	//
	// Experimental.
	MetricCount(props *awscloudwatch.MetricOptions) awscloudwatch.Metric
	// Creates an EventBridge rule for agent events.
	//
	// Returns: An EventBridge Rule configured for agent events.
	// Default: - Default event pattern:
	// - source: ['aws.bedrock']
	// - detail: { 'agent-id': [this.agentId] }
	//
	// Experimental.
	OnEvent(id *string, options *awsevents.OnEventOptions) awsevents.Rule
	// Returns a string representation of this construct.
	// Experimental.
	ToString() *string
	// Applies one or more mixins to this construct.
	//
	// Mixins are applied in order. The list of constructs is captured at the
	// start of the call, so constructs added by a mixin will not be visited.
	// Use multiple `with()` calls if subsequent mixins should apply to added
	// constructs.
	// Experimental.
	With(mixins ...constructs.IMixin) constructs.IConstruct
}

Class to create (or import) an Agent with CDK.

Example: 

parserFunction := lambda.NewFunction(this, jsii.String("ParserFunction"), &FunctionProps{
	Runtime: lambda.Runtime_PYTHON_3_10(),
	Handler: jsii.String("index.handler"),
	Code: lambda.Code_FromAsset(jsii.String("lambda")),
})

agent := bedrock.NewAgent(this, jsii.String("Agent"), &AgentProps{
	FoundationModel: bedrock.BedrockFoundationModel_AMAZON_NOVA_LITE_V1(),
	Instruction: jsii.String("You are a helpful assistant."),
	PromptOverrideConfiguration: bedrock.PromptOverrideConfiguration_WithCustomParser(&CustomParserProps{
		Parser: parserFunction,
		PreProcessingStep: &PromptPreProcessingConfigCustomParser{
			StepType: bedrock.AgentStepType_PRE_PROCESSING,
			UseCustomParser: jsii.Boolean(true),
		},
	}),
})

Experimental.

func NewAgent 

func NewAgent(scope constructs.Construct, id *string, props *AgentProps) Agent

Experimental.

type AgentActionGroup 

type AgentActionGroup interface {
	// The api schema for this action group (if defined).
	// Experimental.
	ApiSchema() ApiSchema
	// A description of the action group.
	// Experimental.
	Description() *string
	// Whether this action group is available for the agent to invoke or not.
	// Experimental.
	Enabled() *bool
	// The action group executor for this action group (if defined).
	// Experimental.
	Executor() ActionGroupExecutor
	// Whether to delete the resource even if it's in use.
	// Experimental.
	ForceDelete() *bool
	// The function schema for this action group (if defined).
	// Experimental.
	FunctionSchema() FunctionSchema
	// The name of the action group.
	// Experimental.
	Name() *string
	// The AWS Defined signature (if defined).
	// Experimental.
	ParentActionGroupSignature() ParentActionGroupSignature
}

**************************************************************************** DEF - Action Group Class ***************************************************************************.

Example: 

bucket := s3.Bucket_FromBucketName(this, jsii.String("ExistingBucket"), jsii.String("my-schema-bucket"))
s3Schema := bedrock.ApiSchema_FromS3File(bucket, jsii.String("schemas/action-group.yaml"))

actionGroupFunction := lambda.NewFunction(this, jsii.String("ActionGroupFunction"), &FunctionProps{
	Runtime: lambda.Runtime_PYTHON_3_12(),
	Handler: jsii.String("index.handler"),
	Code: lambda.Code_FromAsset(path.join(__dirname, jsii.String("../lambda/action-group"))),
})

actionGroup := bedrock.NewAgentActionGroup(&AgentActionGroupProps{
	Name: jsii.String("query-library"),
	Description: jsii.String("Use these functions to get information about the books in the library."),
	Executor: bedrock.ActionGroupExecutor_FromLambda(actionGroupFunction),
	Enabled: jsii.Boolean(true),
	ApiSchema: s3Schema,
})

agent := bedrock.NewAgent(this, jsii.String("Agent"), &AgentProps{
	FoundationModel: bedrock.BedrockFoundationModel_ANTHROPIC_CLAUDE_HAIKU_V1_0(),
	Instruction: jsii.String("You are a helpful and friendly agent that answers questions about literature."),
})

agent.AddActionGroup(actionGroup)

Experimental.

func AgentActionGroup_CodeInterpreter 

func AgentActionGroup_CodeInterpreter(enabled *bool) AgentActionGroup

Defines an action group that allows your agent to request the user for additional information when trying to complete a task. Experimental.

func AgentActionGroup_UserInput 

func AgentActionGroup_UserInput(enabled *bool) AgentActionGroup

Defines an action group that allows your agent to request the user for additional information when trying to complete a task. Experimental.

func NewAgentActionGroup 

func NewAgentActionGroup(props *AgentActionGroupProps) AgentActionGroup

Experimental.

type AgentActionGroupProps 

type AgentActionGroupProps struct {
	// The API Schema defining the functions available to the agent.
	// Default: undefined - No API Schema is provided.
	//
	// Experimental.
	ApiSchema ApiSchema `field:"optional" json:"apiSchema" yaml:"apiSchema"`
	// A description of the action group.
	// Default: undefined - No description is provided.
	//
	// Experimental.
	Description *string `field:"optional" json:"description" yaml:"description"`
	// Specifies whether the action group is available for the agent to invoke or not when sending an InvokeAgent request.
	// Default: true - The action group is enabled.
	//
	// Experimental.
	Enabled *bool `field:"optional" json:"enabled" yaml:"enabled"`
	// The action group executor that implements the API functions.
	// Default: undefined - No executor is provided.
	//
	// Experimental.
	Executor ActionGroupExecutor `field:"optional" json:"executor" yaml:"executor"`
	// Specifies whether to delete the resource even if it's in use.
	// Default: false - The resource will not be deleted if it's in use.
	//
	// Experimental.
	ForceDelete *bool `field:"optional" json:"forceDelete" yaml:"forceDelete"`
	// Defines functions that each define parameters that the agent needs to invoke from the user.
	//
	// NO L2 yet as this doesn't make much sense IMHO.
	// Default: undefined - No function schema is provided.
	//
	// Experimental.
	FunctionSchema FunctionSchema `field:"optional" json:"functionSchema" yaml:"functionSchema"`
	// The name of the action group.
	// Default: - A unique name is generated in the format 'action_group_quick_start_UUID'.
	//
	// Experimental.
	Name *string `field:"optional" json:"name" yaml:"name"`
	// The AWS Defined signature for enabling certain capabilities in your agent.
	//
	// When this property is specified, you must leave the description, apiSchema,
	// and actionGroupExecutor fields blank for this action group.
	// Default: undefined - No parent action group signature is provided.
	//
	// Experimental.
	ParentActionGroupSignature ParentActionGroupSignature `field:"optional" json:"parentActionGroupSignature" yaml:"parentActionGroupSignature"`
}

**************************************************************************** PROPS - Action Group Class ***************************************************************************.

Example: 

bucket := s3.Bucket_FromBucketName(this, jsii.String("ExistingBucket"), jsii.String("my-schema-bucket"))
s3Schema := bedrock.ApiSchema_FromS3File(bucket, jsii.String("schemas/action-group.yaml"))

actionGroupFunction := lambda.NewFunction(this, jsii.String("ActionGroupFunction"), &FunctionProps{
	Runtime: lambda.Runtime_PYTHON_3_12(),
	Handler: jsii.String("index.handler"),
	Code: lambda.Code_FromAsset(path.join(__dirname, jsii.String("../lambda/action-group"))),
})

actionGroup := bedrock.NewAgentActionGroup(&AgentActionGroupProps{
	Name: jsii.String("query-library"),
	Description: jsii.String("Use these functions to get information about the books in the library."),
	Executor: bedrock.ActionGroupExecutor_FromLambda(actionGroupFunction),
	Enabled: jsii.Boolean(true),
	ApiSchema: s3Schema,
})

agent := bedrock.NewAgent(this, jsii.String("Agent"), &AgentProps{
	FoundationModel: bedrock.BedrockFoundationModel_ANTHROPIC_CLAUDE_HAIKU_V1_0(),
	Instruction: jsii.String("You are a helpful and friendly agent that answers questions about literature."),
})

agent.AddActionGroup(actionGroup)

Experimental.

type AgentAlias 

type AgentAlias interface {
	AgentAliasBase
	// The underlying agent for this alias.
	// Experimental.
	Agent() IAgent
	// The ARN of the agent alias.
	// Experimental.
	AliasArn() *string
	// The unique identifier of the agent alias.
	// Experimental.
	AliasId() *string
	// The name of the agent alias.
	//
	// This is either provided by the user or generated from a hash.
	// Experimental.
	AliasName() *string
	// The environment this resource belongs to.
	//
	// For resources that are created and managed in a Stack (those created by
	// creating new class instances like `new Role()`, `new Bucket()`, etc.), this
	// is always the same as the environment of the stack they belong to.
	//
	// For referenced resources (those obtained from referencing methods like
	// `Role.fromRoleArn()`, `Bucket.fromBucketName()`, etc.), they might be
	// different than the stack they were imported into.
	// Experimental.
	Env() *interfaces.ResourceEnvironment
	// The tree node.
	// Experimental.
	Node() constructs.Node
	// Returns a string-encoded token that resolves to the physical name that should be passed to the CloudFormation resource.
	//
	// This value will resolve to one of the following:
	// - a concrete value (e.g. `"my-awesome-bucket"`)
	// - `undefined`, when a name should be generated by CloudFormation
	// - a concrete name generated automatically during synthesis, in
	//   cross-environment scenarios.
	// Experimental.
	PhysicalName() *string
	// The stack in which this resource is defined.
	// Experimental.
	Stack() awscdk.Stack
	// Apply the given removal policy to this resource.
	//
	// The Removal Policy controls what happens to this resource when it stops
	// being managed by CloudFormation, either because you've removed it from the
	// CDK application or because you've made a change that requires the resource
	// to be replaced.
	//
	// The resource can be deleted (`RemovalPolicy.DESTROY`), or left in your AWS
	// account for data recovery and cleanup later (`RemovalPolicy.RETAIN`).
	// Experimental.
	ApplyRemovalPolicy(policy awscdk.RemovalPolicy)
	// Experimental.
	GeneratePhysicalName() *string
	// Returns an environment-sensitive token that should be used for the resource's "ARN" attribute (e.g. `bucket.bucketArn`).
	//
	// Normally, this token will resolve to `arnAttr`, but if the resource is
	// referenced across environments, `arnComponents` will be used to synthesize
	// a concrete ARN with the resource's physical name. Make sure to reference
	// `this.physicalName` in `arnComponents`.
	// Experimental.
	GetResourceArnAttribute(arnAttr *string, arnComponents *awscdk.ArnComponents) *string
	// Returns an environment-sensitive token that should be used for the resource's "name" attribute (e.g. `bucket.bucketName`).
	//
	// Normally, this token will resolve to `nameAttr`, but if the resource is
	// referenced across environments, it will be resolved to `this.physicalName`,
	// which will be a concrete name.
	// Experimental.
	GetResourceNameAttribute(nameAttr *string) *string
	// Grant the given principal identity permissions to perform actions on this agent alias.
	//
	// Note: This grant will only work when the grantee is in the same AWS account
	// where the agent alias is defined. Cross-account grant is not supported.
	// [disable-awslint:no-grants].
	// Experimental.
	Grant(grantee awsiam.IGrantable, actions ...*string) awsiam.Grant
	// Grant the given identity permissions to get the agent alias.
	//
	// Note: This grant will only work when the grantee is in the same AWS account
	// where the agent alias is defined. Cross-account agent read is not supported.
	// [disable-awslint:no-grants].
	// Experimental.
	GrantGet(grantee awsiam.IGrantable) awsiam.Grant
	// Grant the given identity permissions to invoke the agent alias.
	//
	// Note: This grant will only work when the grantee is in the same AWS account
	// where the agent alias is defined. Cross-account invocation is not supported.
	// [disable-awslint:no-grants].
	// Experimental.
	GrantInvoke(grantee awsiam.IGrantable) awsiam.Grant
	// Define an EventBridge rule that triggers when something happens to this agent alias.
	//
	// Requires that there exists at least one CloudTrail Trail in your account
	// that captures the event. This method will not create the Trail.
	// Experimental.
	OnCloudTrailEvent(id *string, options *awsevents.OnEventOptions) awsevents.Rule
	// Returns a string representation of this construct.
	// Experimental.
	ToString() *string
	// Applies one or more mixins to this construct.
	//
	// Mixins are applied in order. The list of constructs is captured at the
	// start of the call, so constructs added by a mixin will not be visited.
	// Use multiple `with()` calls if subsequent mixins should apply to added
	// constructs.
	// Experimental.
	With(mixins ...constructs.IMixin) constructs.IConstruct
}

Class to create an Agent Alias with CDK.

Example: 

// Create a specialized agent
customerSupportAgent := bedrock.NewAgent(this, jsii.String("CustomerSupportAgent"), &AgentProps{
	Instruction: jsii.String("You specialize in answering customer support questions."),
	FoundationModel: bedrock.BedrockFoundationModel_AMAZON_NOVA_LITE_V1(),
})

// Create an agent alias
customerSupportAlias := bedrock.NewAgentAlias(this, jsii.String("CustomerSupportAlias"), &AgentAliasProps{
	Agent: customerSupportAgent,
	AgentAliasName: jsii.String("production"),
})

// Create a main agent that collaborates with the specialized agent
mainAgent := bedrock.NewAgent(this, jsii.String("MainAgent"), &AgentProps{
	Instruction: jsii.String("You route specialized questions to other agents."),
	FoundationModel: bedrock.BedrockFoundationModel_AMAZON_NOVA_LITE_V1(),
	AgentCollaboration: map[string]interface{}{
		"type": bedrock.AgentCollaboratorType_SUPERVISOR,
		"collaborators": []AgentCollaborator{
			bedrock.NewAgentCollaborator(&AgentCollaboratorProps{
				"agentAlias": customerSupportAlias,
				"collaborationInstruction": jsii.String("Route customer support questions to this agent."),
				"collaboratorName": jsii.String("CustomerSupport"),
				"relayConversationHistory": jsii.Boolean(true),
			}),
		},
	},
})

Experimental.

func NewAgentAlias 

func NewAgentAlias(scope constructs.Construct, id *string, props *AgentAliasProps) AgentAlias

Experimental.

type AgentAliasAttributes 

type AgentAliasAttributes struct {
	// The underlying agent for this alias.
	// Experimental.
	Agent IAgent `field:"required" json:"agent" yaml:"agent"`
	// The agent version for this alias.
	// Experimental.
	AgentVersion *string `field:"required" json:"agentVersion" yaml:"agentVersion"`
	// The Id of the agent alias.
	// Experimental.
	AliasId *string `field:"required" json:"aliasId" yaml:"aliasId"`
	// The name of the agent alias.
	// Default: undefined - No alias name is provided.
	//
	// Experimental.
	AliasName *string `field:"optional" json:"aliasName" yaml:"aliasName"`
}

Attributes needed to create an import.

Example: 

cmk := kms.NewKey(this, jsii.String("cmk"), &KeyProps{
})

// Assuming you have an existing agent and alias
agent := bedrock.Agent_FromAgentAttributes(this, jsii.String("ImportedAgent"), &AgentAttributes{
	AgentArn: jsii.String("arn:aws:bedrock:region:account:agent/agent-id"),
	RoleArn: jsii.String("arn:aws:iam::account:role/agent-role"),
})

agentAlias := bedrock.AgentAlias_FromAttributes(this, jsii.String("ImportedAlias"), &AgentAliasAttributes{
	AliasId: jsii.String("alias-id"),
	AliasName: jsii.String("my-alias"),
	AgentVersion: jsii.String("1"),
	Agent: agent,
})

agentVariant := bedrock.PromptVariant_Agent(&AgentPromptVariantProps{
	VariantName: jsii.String("agent-variant"),
	Model: bedrock.BedrockFoundationModel_ANTHROPIC_CLAUDE_3_5_SONNET_V1_0(),
	AgentAlias: agentAlias,
	PromptText: jsii.String("Use the agent to help with: {{task}}. Please be thorough and provide detailed explanations."),
	PromptVariables: []*string{
		jsii.String("task"),
	},
})

bedrock.NewPrompt(this, jsii.String("agentPrompt"), &PromptProps{
	PromptName: jsii.String("agent-prompt"),
	Description: jsii.String("Prompt for agent interactions"),
	DefaultVariant: agentVariant,
	Variants: []IPromptVariant{
		agentVariant,
	},
	KmsKey: cmk,
})

Experimental.

type AgentAliasBase 

type AgentAliasBase interface {
	awscdk.Resource
	IAgentAlias
	// The underlying agent for this alias.
	// Experimental.
	Agent() IAgent
	// The ARN of the agent alias.
	// Experimental.
	AliasArn() *string
	// The unique identifier of the agent alias.
	// Experimental.
	AliasId() *string
	// The environment this resource belongs to.
	//
	// For resources that are created and managed in a Stack (those created by
	// creating new class instances like `new Role()`, `new Bucket()`, etc.), this
	// is always the same as the environment of the stack they belong to.
	//
	// For referenced resources (those obtained from referencing methods like
	// `Role.fromRoleArn()`, `Bucket.fromBucketName()`, etc.), they might be
	// different than the stack they were imported into.
	// Experimental.
	Env() *interfaces.ResourceEnvironment
	// The tree node.
	// Experimental.
	Node() constructs.Node
	// Returns a string-encoded token that resolves to the physical name that should be passed to the CloudFormation resource.
	//
	// This value will resolve to one of the following:
	// - a concrete value (e.g. `"my-awesome-bucket"`)
	// - `undefined`, when a name should be generated by CloudFormation
	// - a concrete name generated automatically during synthesis, in
	//   cross-environment scenarios.
	// Experimental.
	PhysicalName() *string
	// The stack in which this resource is defined.
	// Experimental.
	Stack() awscdk.Stack
	// Apply the given removal policy to this resource.
	//
	// The Removal Policy controls what happens to this resource when it stops
	// being managed by CloudFormation, either because you've removed it from the
	// CDK application or because you've made a change that requires the resource
	// to be replaced.
	//
	// The resource can be deleted (`RemovalPolicy.DESTROY`), or left in your AWS
	// account for data recovery and cleanup later (`RemovalPolicy.RETAIN`).
	// Experimental.
	ApplyRemovalPolicy(policy awscdk.RemovalPolicy)
	// Experimental.
	GeneratePhysicalName() *string
	// Returns an environment-sensitive token that should be used for the resource's "ARN" attribute (e.g. `bucket.bucketArn`).
	//
	// Normally, this token will resolve to `arnAttr`, but if the resource is
	// referenced across environments, `arnComponents` will be used to synthesize
	// a concrete ARN with the resource's physical name. Make sure to reference
	// `this.physicalName` in `arnComponents`.
	// Experimental.
	GetResourceArnAttribute(arnAttr *string, arnComponents *awscdk.ArnComponents) *string
	// Returns an environment-sensitive token that should be used for the resource's "name" attribute (e.g. `bucket.bucketName`).
	//
	// Normally, this token will resolve to `nameAttr`, but if the resource is
	// referenced across environments, it will be resolved to `this.physicalName`,
	// which will be a concrete name.
	// Experimental.
	GetResourceNameAttribute(nameAttr *string) *string
	// Grant the given principal identity permissions to perform actions on this agent alias.
	//
	// Note: This grant will only work when the grantee is in the same AWS account
	// where the agent alias is defined. Cross-account grant is not supported.
	// [disable-awslint:no-grants].
	// Experimental.
	Grant(grantee awsiam.IGrantable, actions ...*string) awsiam.Grant
	// Grant the given identity permissions to get the agent alias.
	//
	// Note: This grant will only work when the grantee is in the same AWS account
	// where the agent alias is defined. Cross-account agent read is not supported.
	// [disable-awslint:no-grants].
	// Experimental.
	GrantGet(grantee awsiam.IGrantable) awsiam.Grant
	// Grant the given identity permissions to invoke the agent alias.
	//
	// Note: This grant will only work when the grantee is in the same AWS account
	// where the agent alias is defined. Cross-account invocation is not supported.
	// [disable-awslint:no-grants].
	// Experimental.
	GrantInvoke(grantee awsiam.IGrantable) awsiam.Grant
	// Define an EventBridge rule that triggers when something happens to this agent alias.
	//
	// Requires that there exists at least one CloudTrail Trail in your account
	// that captures the event. This method will not create the Trail.
	// Experimental.
	OnCloudTrailEvent(id *string, options *awsevents.OnEventOptions) awsevents.Rule
	// Returns a string representation of this construct.
	// Experimental.
	ToString() *string
	// Applies one or more mixins to this construct.
	//
	// Mixins are applied in order. The list of constructs is captured at the
	// start of the call, so constructs added by a mixin will not be visited.
	// Use multiple `with()` calls if subsequent mixins should apply to added
	// constructs.
	// Experimental.
	With(mixins ...constructs.IMixin) constructs.IConstruct
}

Abstract base class for an Agent.

Contains methods and attributes valid for Agents either created with CDK or imported. Experimental.

type AgentAliasProps 

type AgentAliasProps struct {
	// The agent associated to this alias.
	// Experimental.
	Agent IAgent `field:"required" json:"agent" yaml:"agent"`
	// The name for the agent alias.
	//
	// This will be used as the physical name of the agent alias.
	// Default: "latest".
	//
	// Experimental.
	AgentAliasName *string `field:"optional" json:"agentAliasName" yaml:"agentAliasName"`
	// The version of the agent to associate with the agent alias.
	// Default: - Creates a new version of the agent.
	//
	// Experimental.
	AgentVersion *string `field:"optional" json:"agentVersion" yaml:"agentVersion"`
	// Description for the agent alias.
	// Default: undefined - No description is provided.
	//
	// Experimental.
	Description *string `field:"optional" json:"description" yaml:"description"`
}

Properties for creating a CDK-Managed Agent Alias.

Example: 

// Create a specialized agent
customerSupportAgent := bedrock.NewAgent(this, jsii.String("CustomerSupportAgent"), &AgentProps{
	Instruction: jsii.String("You specialize in answering customer support questions."),
	FoundationModel: bedrock.BedrockFoundationModel_AMAZON_NOVA_LITE_V1(),
})

// Create an agent alias
customerSupportAlias := bedrock.NewAgentAlias(this, jsii.String("CustomerSupportAlias"), &AgentAliasProps{
	Agent: customerSupportAgent,
	AgentAliasName: jsii.String("production"),
})

// Create a main agent that collaborates with the specialized agent
mainAgent := bedrock.NewAgent(this, jsii.String("MainAgent"), &AgentProps{
	Instruction: jsii.String("You route specialized questions to other agents."),
	FoundationModel: bedrock.BedrockFoundationModel_AMAZON_NOVA_LITE_V1(),
	AgentCollaboration: map[string]interface{}{
		"type": bedrock.AgentCollaboratorType_SUPERVISOR,
		"collaborators": []AgentCollaborator{
			bedrock.NewAgentCollaborator(&AgentCollaboratorProps{
				"agentAlias": customerSupportAlias,
				"collaborationInstruction": jsii.String("Route customer support questions to this agent."),
				"collaboratorName": jsii.String("CustomerSupport"),
				"relayConversationHistory": jsii.Boolean(true),
			}),
		},
	},
})

Experimental.

type AgentAttributes 

type AgentAttributes struct {
	// The ARN of the agent.
	// Experimental.
	AgentArn *string `field:"required" json:"agentArn" yaml:"agentArn"`
	// The ARN of the IAM role associated to the agent.
	// Experimental.
	RoleArn *string `field:"required" json:"roleArn" yaml:"roleArn"`
	// The agent version.
	//
	// If no explicit versions have been created,
	// leave this empty to use the DRAFT version. Otherwise, use the
	// version number (e.g. 1).
	// Default: 'DRAFT'.
	//
	// Experimental.
	AgentVersion *string `field:"optional" json:"agentVersion" yaml:"agentVersion"`
	// Optional KMS encryption key associated with this agent.
	// Default: undefined - An AWS managed key is used.
	//
	// Experimental.
	KmsKeyArn *string `field:"optional" json:"kmsKeyArn" yaml:"kmsKeyArn"`
	// When this agent was last updated.
	// Default: undefined - No last updated timestamp is provided.
	//
	// Experimental.
	LastUpdated *string `field:"optional" json:"lastUpdated" yaml:"lastUpdated"`
}

Attributes for specifying an imported Bedrock Agent.

Example: 

cmk := kms.NewKey(this, jsii.String("cmk"), &KeyProps{
})

// Assuming you have an existing agent and alias
agent := bedrock.Agent_FromAgentAttributes(this, jsii.String("ImportedAgent"), &AgentAttributes{
	AgentArn: jsii.String("arn:aws:bedrock:region:account:agent/agent-id"),
	RoleArn: jsii.String("arn:aws:iam::account:role/agent-role"),
})

agentAlias := bedrock.AgentAlias_FromAttributes(this, jsii.String("ImportedAlias"), &AgentAliasAttributes{
	AliasId: jsii.String("alias-id"),
	AliasName: jsii.String("my-alias"),
	AgentVersion: jsii.String("1"),
	Agent: agent,
})

agentVariant := bedrock.PromptVariant_Agent(&AgentPromptVariantProps{
	VariantName: jsii.String("agent-variant"),
	Model: bedrock.BedrockFoundationModel_ANTHROPIC_CLAUDE_3_5_SONNET_V1_0(),
	AgentAlias: agentAlias,
	PromptText: jsii.String("Use the agent to help with: {{task}}. Please be thorough and provide detailed explanations."),
	PromptVariables: []*string{
		jsii.String("task"),
	},
})

bedrock.NewPrompt(this, jsii.String("agentPrompt"), &PromptProps{
	PromptName: jsii.String("agent-prompt"),
	Description: jsii.String("Prompt for agent interactions"),
	DefaultVariant: agentVariant,
	Variants: []IPromptVariant{
		agentVariant,
	},
	KmsKey: cmk,
})

Experimental.

type AgentBase 

type AgentBase interface {
	awscdk.Resource
	IAgent
	// The ARN of the agent.
	// Experimental.
	AgentArn() *string
	// The ID of the Agent.
	// Experimental.
	AgentId() *string
	// The version of the agent.
	// Experimental.
	AgentVersion() *string
	// The environment this resource belongs to.
	//
	// For resources that are created and managed in a Stack (those created by
	// creating new class instances like `new Role()`, `new Bucket()`, etc.), this
	// is always the same as the environment of the stack they belong to.
	//
	// For referenced resources (those obtained from referencing methods like
	// `Role.fromRoleArn()`, `Bucket.fromBucketName()`, etc.), they might be
	// different than the stack they were imported into.
	// Experimental.
	Env() *interfaces.ResourceEnvironment
	// The principal to grant permissions to.
	// Experimental.
	GrantPrincipal() awsiam.IPrincipal
	// Optional KMS encryption key associated with this agent.
	// Experimental.
	KmsKey() awskms.IKey
	// When this agent was last updated.
	// Experimental.
	LastUpdated() *string
	// The tree node.
	// Experimental.
	Node() constructs.Node
	// Returns a string-encoded token that resolves to the physical name that should be passed to the CloudFormation resource.
	//
	// This value will resolve to one of the following:
	// - a concrete value (e.g. `"my-awesome-bucket"`)
	// - `undefined`, when a name should be generated by CloudFormation
	// - a concrete name generated automatically during synthesis, in
	//   cross-environment scenarios.
	// Experimental.
	PhysicalName() *string
	// The IAM role associated to the agent.
	// Experimental.
	Role() awsiam.IRole
	// The stack in which this resource is defined.
	// Experimental.
	Stack() awscdk.Stack
	// Apply the given removal policy to this resource.
	//
	// The Removal Policy controls what happens to this resource when it stops
	// being managed by CloudFormation, either because you've removed it from the
	// CDK application or because you've made a change that requires the resource
	// to be replaced.
	//
	// The resource can be deleted (`RemovalPolicy.DESTROY`), or left in your AWS
	// account for data recovery and cleanup later (`RemovalPolicy.RETAIN`).
	// Experimental.
	ApplyRemovalPolicy(policy awscdk.RemovalPolicy)
	// Experimental.
	GeneratePhysicalName() *string
	// Returns an environment-sensitive token that should be used for the resource's "ARN" attribute (e.g. `bucket.bucketArn`).
	//
	// Normally, this token will resolve to `arnAttr`, but if the resource is
	// referenced across environments, `arnComponents` will be used to synthesize
	// a concrete ARN with the resource's physical name. Make sure to reference
	// `this.physicalName` in `arnComponents`.
	// Experimental.
	GetResourceArnAttribute(arnAttr *string, arnComponents *awscdk.ArnComponents) *string
	// Returns an environment-sensitive token that should be used for the resource's "name" attribute (e.g. `bucket.bucketName`).
	//
	// Normally, this token will resolve to `nameAttr`, but if the resource is
	// referenced across environments, it will be resolved to `this.physicalName`,
	// which will be a concrete name.
	// Experimental.
	GetResourceNameAttribute(nameAttr *string) *string
	// Grant invoke permissions on this agent to an IAM principal.
	//
	// [disable-awslint:no-grants].
	//
	// Returns: An IAM Grant object representing the granted permissions.
	// Default: - Default grant configuration:
	// - actions: ['bedrock:InvokeAgent']
	// - resourceArns: [this.agentArn]
	//
	// Experimental.
	GrantInvoke(grantee awsiam.IGrantable) awsiam.Grant
	// Creates a CloudWatch metric for tracking agent invocations.
	//
	// Returns: A CloudWatch Metric configured for agent invocation counts.
	// Default: - Default metric configuration:
	// - namespace: 'AWS/Bedrock'
	// - metricName: 'Invocations'
	// - dimensionsMap: { AgentId: this.agentId }
	//
	// Experimental.
	MetricCount(props *awscloudwatch.MetricOptions) awscloudwatch.Metric
	// Creates an EventBridge rule for agent events.
	//
	// Returns: An EventBridge Rule configured for agent events.
	// Default: - Default event pattern:
	// - source: ['aws.bedrock']
	// - detail: { 'agent-id': [this.agentId] }
	//
	// Experimental.
	OnEvent(id *string, options *awsevents.OnEventOptions) awsevents.Rule
	// Returns a string representation of this construct.
	// Experimental.
	ToString() *string
	// Applies one or more mixins to this construct.
	//
	// Mixins are applied in order. The list of constructs is captured at the
	// start of the call, so constructs added by a mixin will not be visited.
	// Use multiple `with()` calls if subsequent mixins should apply to added
	// constructs.
	// Experimental.
	With(mixins ...constructs.IMixin) constructs.IConstruct
}

Abstract base class for an Agent.

Contains methods and attributes valid for Agents either created with CDK or imported. Experimental.

type AgentCollaboration 

type AgentCollaboration interface {
	// Collaborators that this agent will work with.
	// Experimental.
	Collaborators() *[]AgentCollaborator
	// The collaboration type for the agent.
	// Experimental.
	Type() AgentCollaboratorType
}

Class to manage agent collaboration configuration.

Example: 

// Create a specialized agent
customerSupportAgent := bedrock.NewAgent(this, jsii.String("CustomerSupportAgent"), &AgentProps{
	Instruction: jsii.String("You specialize in answering customer support questions."),
	FoundationModel: bedrock.BedrockFoundationModel_AMAZON_NOVA_LITE_V1(),
})

// Create an agent alias
customerSupportAlias := bedrock.NewAgentAlias(this, jsii.String("CustomerSupportAlias"), &AgentAliasProps{
	Agent: customerSupportAgent,
	AgentAliasName: jsii.String("production"),
})

// Create a main agent that collaborates with the specialized agent
mainAgent := bedrock.NewAgent(this, jsii.String("MainAgent"), &AgentProps{
	Instruction: jsii.String("You route specialized questions to other agents."),
	FoundationModel: bedrock.BedrockFoundationModel_AMAZON_NOVA_LITE_V1(),
	AgentCollaboration: map[string]interface{}{
		"type": bedrock.AgentCollaboratorType_SUPERVISOR,
		"collaborators": []AgentCollaborator{
			bedrock.NewAgentCollaborator(&AgentCollaboratorProps{
				"agentAlias": customerSupportAlias,
				"collaborationInstruction": jsii.String("Route customer support questions to this agent."),
				"collaboratorName": jsii.String("CustomerSupport"),
				"relayConversationHistory": jsii.Boolean(true),
			}),
		},
	},
})

Experimental.

func NewAgentCollaboration 

func NewAgentCollaboration(config *AgentCollaborationConfig) AgentCollaboration

Experimental.

type AgentCollaborationConfig 

type AgentCollaborationConfig struct {
	// Collaborators that this agent will work with.
	// Experimental.
	Collaborators *[]AgentCollaborator `field:"required" json:"collaborators" yaml:"collaborators"`
	// The collaboration type for the agent.
	// Experimental.
	Type AgentCollaboratorType `field:"required" json:"type" yaml:"type"`
}

Configuration for agent collaboration settings.

Example: 

// The code below shows an example of how to instantiate this type.
// The values are placeholders you should change.
import bedrock_alpha "github.com/aws/aws-cdk-go/awsbedrockalpha"

var agentCollaborator AgentCollaborator

agentCollaborationConfig := &AgentCollaborationConfig{
	Collaborators: []AgentCollaborator{
		agentCollaborator,
	},
	Type: bedrock_alpha.AgentCollaboratorType_SUPERVISOR,
}

Experimental.

type AgentCollaborator 

type AgentCollaborator interface {
	// The agent alias that this collaborator represents.
	//
	// This is the agent that will be called upon for collaboration.
	// Experimental.
	AgentAlias() IAgentAlias
	// Instructions on how this agent should collaborate with the main agent.
	// Experimental.
	CollaborationInstruction() *string
	// A friendly name for the collaborator.
	// Experimental.
	CollaboratorName() *string
	// Whether to relay conversation history to this collaborator.
	// Default: - undefined (uses service default).
	//
	// Experimental.
	RelayConversationHistory() *bool
	// Grants the given identity permissions to collaborate with the agent [disable-awslint:no-grants].
	//
	// Returns: The Grant object.
	// Experimental.
	Grant(grantee awsiam.IGrantable) awsiam.Grant
}

**************************************************************************** Agent Collaborator Class ***************************************************************************.

Example: 

// Create a specialized agent
customerSupportAgent := bedrock.NewAgent(this, jsii.String("CustomerSupportAgent"), &AgentProps{
	Instruction: jsii.String("You specialize in answering customer support questions."),
	FoundationModel: bedrock.BedrockFoundationModel_AMAZON_NOVA_LITE_V1(),
})

// Create an agent alias
customerSupportAlias := bedrock.NewAgentAlias(this, jsii.String("CustomerSupportAlias"), &AgentAliasProps{
	Agent: customerSupportAgent,
	AgentAliasName: jsii.String("production"),
})

// Create a main agent that collaborates with the specialized agent
mainAgent := bedrock.NewAgent(this, jsii.String("MainAgent"), &AgentProps{
	Instruction: jsii.String("You route specialized questions to other agents."),
	FoundationModel: bedrock.BedrockFoundationModel_AMAZON_NOVA_LITE_V1(),
	AgentCollaboration: map[string]interface{}{
		"type": bedrock.AgentCollaboratorType_SUPERVISOR,
		"collaborators": []AgentCollaborator{
			bedrock.NewAgentCollaborator(&AgentCollaboratorProps{
				"agentAlias": customerSupportAlias,
				"collaborationInstruction": jsii.String("Route customer support questions to this agent."),
				"collaboratorName": jsii.String("CustomerSupport"),
				"relayConversationHistory": jsii.Boolean(true),
			}),
		},
	},
})

Experimental.

func NewAgentCollaborator 

func NewAgentCollaborator(props *AgentCollaboratorProps) AgentCollaborator

Experimental.

type AgentCollaboratorProps 

type AgentCollaboratorProps struct {
	// Descriptor for the collaborating agent.
	//
	// This cannot be the TSTALIASID (`agent.testAlias`).
	// Experimental.
	AgentAlias IAgentAlias `field:"required" json:"agentAlias" yaml:"agentAlias"`
	// Instructions on how this agent should collaborate with the main agent.
	// Experimental.
	CollaborationInstruction *string `field:"required" json:"collaborationInstruction" yaml:"collaborationInstruction"`
	// A friendly name for the collaborator.
	// Experimental.
	CollaboratorName *string `field:"required" json:"collaboratorName" yaml:"collaboratorName"`
	// Whether to relay conversation history to this collaborator.
	// Default: - undefined (uses service default).
	//
	// Experimental.
	RelayConversationHistory *bool `field:"optional" json:"relayConversationHistory" yaml:"relayConversationHistory"`
}

**************************************************************************** PROPS - Agent Collaborator Class ***************************************************************************.

Example: 

// Create a specialized agent
customerSupportAgent := bedrock.NewAgent(this, jsii.String("CustomerSupportAgent"), &AgentProps{
	Instruction: jsii.String("You specialize in answering customer support questions."),
	FoundationModel: bedrock.BedrockFoundationModel_AMAZON_NOVA_LITE_V1(),
})

// Create an agent alias
customerSupportAlias := bedrock.NewAgentAlias(this, jsii.String("CustomerSupportAlias"), &AgentAliasProps{
	Agent: customerSupportAgent,
	AgentAliasName: jsii.String("production"),
})

// Create a main agent that collaborates with the specialized agent
mainAgent := bedrock.NewAgent(this, jsii.String("MainAgent"), &AgentProps{
	Instruction: jsii.String("You route specialized questions to other agents."),
	FoundationModel: bedrock.BedrockFoundationModel_AMAZON_NOVA_LITE_V1(),
	AgentCollaboration: map[string]interface{}{
		"type": bedrock.AgentCollaboratorType_SUPERVISOR,
		"collaborators": []AgentCollaborator{
			bedrock.NewAgentCollaborator(&AgentCollaboratorProps{
				"agentAlias": customerSupportAlias,
				"collaborationInstruction": jsii.String("Route customer support questions to this agent."),
				"collaboratorName": jsii.String("CustomerSupport"),
				"relayConversationHistory": jsii.Boolean(true),
			}),
		},
	},
})

Experimental.

type AgentCollaboratorType 

type AgentCollaboratorType string

Enum for collaborator's relay conversation history types.

Example: 

// Create a specialized agent
customerSupportAgent := bedrock.NewAgent(this, jsii.String("CustomerSupportAgent"), &AgentProps{
	Instruction: jsii.String("You specialize in answering customer support questions."),
	FoundationModel: bedrock.BedrockFoundationModel_AMAZON_NOVA_LITE_V1(),
})

// Create an agent alias
customerSupportAlias := bedrock.NewAgentAlias(this, jsii.String("CustomerSupportAlias"), &AgentAliasProps{
	Agent: customerSupportAgent,
	AgentAliasName: jsii.String("production"),
})

// Create a main agent that collaborates with the specialized agent
mainAgent := bedrock.NewAgent(this, jsii.String("MainAgent"), &AgentProps{
	Instruction: jsii.String("You route specialized questions to other agents."),
	FoundationModel: bedrock.BedrockFoundationModel_AMAZON_NOVA_LITE_V1(),
	AgentCollaboration: map[string]interface{}{
		"type": bedrock.AgentCollaboratorType_SUPERVISOR,
		"collaborators": []AgentCollaborator{
			bedrock.NewAgentCollaborator(&AgentCollaboratorProps{
				"agentAlias": customerSupportAlias,
				"collaborationInstruction": jsii.String("Route customer support questions to this agent."),
				"collaboratorName": jsii.String("CustomerSupport"),
				"relayConversationHistory": jsii.Boolean(true),
			}),
		},
	},
})

Experimental. 

const (
	// Supervisor agent.
	// Experimental.
	AgentCollaboratorType_SUPERVISOR AgentCollaboratorType = "SUPERVISOR"
	// Disabling collaboration.
	// Experimental.
	AgentCollaboratorType_DISABLED AgentCollaboratorType = "DISABLED"
	// Supervisor router.
	// Experimental.
	AgentCollaboratorType_SUPERVISOR_ROUTER AgentCollaboratorType = "SUPERVISOR_ROUTER"
)

type AgentGenAiResourceProps 

type AgentGenAiResourceProps struct {
	// The agent alias to use for the GenAI resource.
	// Experimental.
	AgentAlias IAgentAlias `field:"required" json:"agentAlias" yaml:"agentAlias"`
}

Properties for creating an agent GenAI resource configuration.

Example: 

// The code below shows an example of how to instantiate this type.
// The values are placeholders you should change.
import bedrock_alpha "github.com/aws/aws-cdk-go/awsbedrockalpha"

var agentAlias AgentAlias

agentGenAiResourceProps := &AgentGenAiResourceProps{
	AgentAlias: agentAlias,
}

Experimental.

type AgentPromptVariantProps 

type AgentPromptVariantProps struct {
	// The model which is used to run the prompt.
	//
	// The model could be a foundation
	// model, a custom model, or a provisioned model.
	// Experimental.
	Model IBedrockInvokable `field:"required" json:"model" yaml:"model"`
	// The name of the prompt variant.
	// Experimental.
	VariantName *string `field:"required" json:"variantName" yaml:"variantName"`
	// The variables in the prompt template that can be filled in at runtime.
	// Default: - No variables defined.
	//
	// Experimental.
	PromptVariables *[]*string `field:"optional" json:"promptVariables" yaml:"promptVariables"`
	// An alias pointing to the agent version to be used.
	// Experimental.
	AgentAlias IAgentAlias `field:"required" json:"agentAlias" yaml:"agentAlias"`
	// The text prompt.
	//
	// Variables are used by enclosing its name with double curly braces
	// as in `{{variable_name}}`.
	// Experimental.
	PromptText *string `field:"required" json:"promptText" yaml:"promptText"`
}

Properties for creating an agent prompt variant.

Example: 

cmk := kms.NewKey(this, jsii.String("cmk"), &KeyProps{
})

// Assuming you have an existing agent and alias
agent := bedrock.Agent_FromAgentAttributes(this, jsii.String("ImportedAgent"), &AgentAttributes{
	AgentArn: jsii.String("arn:aws:bedrock:region:account:agent/agent-id"),
	RoleArn: jsii.String("arn:aws:iam::account:role/agent-role"),
})

agentAlias := bedrock.AgentAlias_FromAttributes(this, jsii.String("ImportedAlias"), &AgentAliasAttributes{
	AliasId: jsii.String("alias-id"),
	AliasName: jsii.String("my-alias"),
	AgentVersion: jsii.String("1"),
	Agent: agent,
})

agentVariant := bedrock.PromptVariant_Agent(&AgentPromptVariantProps{
	VariantName: jsii.String("agent-variant"),
	Model: bedrock.BedrockFoundationModel_ANTHROPIC_CLAUDE_3_5_SONNET_V1_0(),
	AgentAlias: agentAlias,
	PromptText: jsii.String("Use the agent to help with: {{task}}. Please be thorough and provide detailed explanations."),
	PromptVariables: []*string{
		jsii.String("task"),
	},
})

bedrock.NewPrompt(this, jsii.String("agentPrompt"), &PromptProps{
	PromptName: jsii.String("agent-prompt"),
	Description: jsii.String("Prompt for agent interactions"),
	DefaultVariant: agentVariant,
	Variants: []IPromptVariant{
		agentVariant,
	},
	KmsKey: cmk,
})

Experimental.

type AgentProps 

type AgentProps struct {
	// The foundation model used for orchestration by the agent.
	// Experimental.
	FoundationModel IBedrockInvokable `field:"required" json:"foundationModel" yaml:"foundationModel"`
	// The instruction used by the agent.
	//
	// This determines how the agent will perform his task.
	// This instruction must have a minimum of 40 characters.
	// Experimental.
	Instruction *string `field:"required" json:"instruction" yaml:"instruction"`
	// The Action Groups associated with the agent.
	// Default: - Only default action groups (UserInput and CodeInterpreter) are added.
	//
	// Experimental.
	ActionGroups *[]AgentActionGroup `field:"optional" json:"actionGroups" yaml:"actionGroups"`
	// Configuration for agent collaboration settings, including AgentCollaboratorType and AgentCollaborators.
	//
	// This property allows you to define how the agent collaborates with other agents
	// and what collaborators it can work with.
	// Default: - No agent collaboration configuration.
	//
	// Experimental.
	AgentCollaboration AgentCollaboration `field:"optional" json:"agentCollaboration" yaml:"agentCollaboration"`
	// The name of the agent.
	//
	// This will be used as the physical name of the agent.
	// Default: - A name is generated by CDK.
	// Supported pattern : ^([0-9a-zA-Z][_-]?){1,100}$
	//
	// Experimental.
	AgentName *string `field:"optional" json:"agentName" yaml:"agentName"`
	// Select whether the agent can generate, run, and troubleshoot code when trying to complete a task.
	// Default: false.
	//
	// Experimental.
	CodeInterpreterEnabled *bool `field:"optional" json:"codeInterpreterEnabled" yaml:"codeInterpreterEnabled"`
	// The Lambda function to use for custom orchestration.
	//
	// If provided, custom orchestration will be used.
	// If not provided, default orchestration will be used.
	// Default: - Default orchestration.
	//
	// Experimental.
	CustomOrchestrationExecutor CustomOrchestrationExecutor `field:"optional" json:"customOrchestrationExecutor" yaml:"customOrchestrationExecutor"`
	// A description of the agent.
	// Default: - No description is provided.
	//
	// Experimental.
	Description *string `field:"optional" json:"description" yaml:"description"`
	// An existing IAM Role to associate with this agent.
	//
	// Use this property when you want to reuse an existing IAM role rather than create a new one.
	// The role must have a trust policy that allows the Bedrock service to assume it.
	// Default: - A new role is created for you.
	//
	// Experimental.
	ExistingRole awsiam.IRole `field:"optional" json:"existingRole" yaml:"existingRole"`
	// Whether to delete the resource even if it's in use.
	// Default: false.
	//
	// Experimental.
	ForceDelete *bool `field:"optional" json:"forceDelete" yaml:"forceDelete"`
	// The guardrail that will be associated with the agent.
	// Default: - No guardrail is provided.
	//
	// Experimental.
	Guardrail IGuardrail `field:"optional" json:"guardrail" yaml:"guardrail"`
	// How long sessions should be kept open for the agent.
	//
	// If no conversation occurs
	// during this time, the session expires and Amazon Bedrock deletes any data
	// provided before the timeout.
	// Default: - 10 minutes.
	//
	// Experimental.
	IdleSessionTTL awscdk.Duration `field:"optional" json:"idleSessionTTL" yaml:"idleSessionTTL"`
	// The KMS key of the agent if custom encryption is configured.
	// Default: - An AWS managed key is used.
	//
	// Experimental.
	KmsKey awskms.IKey `field:"optional" json:"kmsKey" yaml:"kmsKey"`
	// The type and configuration of the memory to maintain context across multiple sessions and recall past interactions.
	//
	// This can be useful for maintaining continuity in multi-turn conversations and recalling user preferences
	// or past interactions.
	// See: https://docs.aws.amazon.com/bedrock/latest/userguide/agents-memory.html
	//
	// Default: - No memory will be used. Agents will retain context from the current session only.
	//
	// Experimental.
	Memory Memory `field:"optional" json:"memory" yaml:"memory"`
	// Overrides some prompt templates in different parts of an agent sequence configuration.
	// Default: - No overrides are provided.
	//
	// Experimental.
	PromptOverrideConfiguration PromptOverrideConfiguration `field:"optional" json:"promptOverrideConfiguration" yaml:"promptOverrideConfiguration"`
	// Specifies whether to automatically update the `DRAFT` version of the agent after making changes to the agent.
	//
	// The `DRAFT` version can be continually iterated
	// upon during internal development.
	// Default: false.
	//
	// Experimental.
	ShouldPrepareAgent *bool `field:"optional" json:"shouldPrepareAgent" yaml:"shouldPrepareAgent"`
	// Select whether the agent can prompt additional information from the user when it does not have enough information to respond to an utterance.
	// Default: false.
	//
	// Experimental.
	UserInputEnabled *bool `field:"optional" json:"userInputEnabled" yaml:"userInputEnabled"`
}

Properties for creating a CDK managed Bedrock Agent.

TODO: Knowledge bases configuration will be added in a future update TODO: Inference profile configuration will be added in a future update.

Example: 

parserFunction := lambda.NewFunction(this, jsii.String("ParserFunction"), &FunctionProps{
	Runtime: lambda.Runtime_PYTHON_3_10(),
	Handler: jsii.String("index.handler"),
	Code: lambda.Code_FromAsset(jsii.String("lambda")),
})

agent := bedrock.NewAgent(this, jsii.String("Agent"), &AgentProps{
	FoundationModel: bedrock.BedrockFoundationModel_AMAZON_NOVA_LITE_V1(),
	Instruction: jsii.String("You are a helpful assistant."),
	PromptOverrideConfiguration: bedrock.PromptOverrideConfiguration_WithCustomParser(&CustomParserProps{
		Parser: parserFunction,
		PreProcessingStep: &PromptPreProcessingConfigCustomParser{
			StepType: bedrock.AgentStepType_PRE_PROCESSING,
			UseCustomParser: jsii.Boolean(true),
		},
	}),
})

Experimental.

type AgentStepType 

type AgentStepType string

The step in the agent sequence that this prompt configuration applies to.

Example: 

agent := bedrock.NewAgent(this, jsii.String("Agent"), &AgentProps{
	FoundationModel: bedrock.BedrockFoundationModel_AMAZON_NOVA_LITE_V1(),
	Instruction: jsii.String("You are a helpful assistant."),
	PromptOverrideConfiguration: bedrock.PromptOverrideConfiguration_FromSteps([]PromptStepConfigBase{
		&PromptRoutingClassifierConfigCustomParser{
			StepType: bedrock.AgentStepType_ROUTING_CLASSIFIER,
			StepEnabled: jsii.Boolean(true),
			CustomPromptTemplate: jsii.String("Your routing template here"),
			FoundationModel: bedrock.BedrockFoundationModel_ANTHROPIC_CLAUDE_V2(),
		}.(PromptRoutingClassifierConfigCustomParser),
	}),
})

Experimental. 

const (
	// Pre-processing step that prepares the user input for orchestration.
	// Experimental.
	AgentStepType_PRE_PROCESSING AgentStepType = "PRE_PROCESSING"
	// Main orchestration step that determines the agent's actions.
	// Experimental.
	AgentStepType_ORCHESTRATION AgentStepType = "ORCHESTRATION"
	// Post-processing step that refines the agent's response.
	// Experimental.
	AgentStepType_POST_PROCESSING AgentStepType = "POST_PROCESSING"
	// Step that classifies and routes requests to appropriate collaborators.
	// Experimental.
	AgentStepType_ROUTING_CLASSIFIER AgentStepType = "ROUTING_CLASSIFIER"
	// Step that summarizes conversation history for memory retention.
	// Experimental.
	AgentStepType_MEMORY_SUMMARIZATION AgentStepType = "MEMORY_SUMMARIZATION"
	// Step that generates responses using knowledge base content.
	// Experimental.
	AgentStepType_KNOWLEDGE_BASE_RESPONSE_GENERATION AgentStepType = "KNOWLEDGE_BASE_RESPONSE_GENERATION"
)

type ApiSchema 

type ApiSchema interface {
	// The inline OpenAPI schema definition as a string, if using an inline schema.
	//
	// Can be in JSON or YAML format.
	// Experimental.
	InlineSchema() *string
	// The S3 location of the API schema file, if using an S3-based schema.
	//
	// Contains the bucket name and object key information.
	// Experimental.
	S3File() *awss3.Location
}

Represents the concept of an API Schema for a Bedrock Agent Action Group.

Example: 

bucket := s3.Bucket_FromBucketName(this, jsii.String("ExistingBucket"), jsii.String("my-schema-bucket"))
s3Schema := bedrock.ApiSchema_FromS3File(bucket, jsii.String("schemas/action-group.yaml"))

actionGroupFunction := lambda.NewFunction(this, jsii.String("ActionGroupFunction"), &FunctionProps{
	Runtime: lambda.Runtime_PYTHON_3_12(),
	Handler: jsii.String("index.handler"),
	Code: lambda.Code_FromAsset(path.join(__dirname, jsii.String("../lambda/action-group"))),
})

actionGroup := bedrock.NewAgentActionGroup(&AgentActionGroupProps{
	Name: jsii.String("query-library"),
	Description: jsii.String("Use these functions to get information about the books in the library."),
	Executor: bedrock.ActionGroupExecutor_FromLambda(actionGroupFunction),
	Enabled: jsii.Boolean(true),
	ApiSchema: s3Schema,
})

agent := bedrock.NewAgent(this, jsii.String("Agent"), &AgentProps{
	FoundationModel: bedrock.BedrockFoundationModel_ANTHROPIC_CLAUDE_HAIKU_V1_0(),
	Instruction: jsii.String("You are a helpful and friendly agent that answers questions about literature."),
})

agent.AddActionGroup(actionGroup)

Experimental.

type ApplicationInferenceProfile 

type ApplicationInferenceProfile interface {
	InferenceProfileBase
	IBedrockInvokable
	// Time Stamp for Application Inference Profile creation.
	// Experimental.
	CreatedAt() *string
	// The environment this resource belongs to.
	//
	// For resources that are created and managed in a Stack (those created by
	// creating new class instances like `new Role()`, `new Bucket()`, etc.), this
	// is always the same as the environment of the stack they belong to.
	//
	// For referenced resources (those obtained from referencing methods like
	// `Role.fromRoleArn()`, `Bucket.fromBucketName()`, etc.), they might be
	// different than the stack they were imported into.
	// Experimental.
	Env() *interfaces.ResourceEnvironment
	// The ARN of the application inference profile.
	// Experimental.
	InferenceProfileArn() *string
	// The unique identifier of the application inference profile.
	// Experimental.
	InferenceProfileId() *string
	// The underlying model/cross-region model used by the application inference profile.
	// Experimental.
	InferenceProfileModel() IBedrockInvokable
	// The name of the application inference profile.
	// Experimental.
	InferenceProfileName() *string
	// The ARN used for invoking this inference profile.
	//
	// This equals to the inferenceProfileArn property, useful for implementing IBedrockInvokable interface.
	// Experimental.
	InvokableArn() *string
	// The tree node.
	// Experimental.
	Node() constructs.Node
	// Returns a string-encoded token that resolves to the physical name that should be passed to the CloudFormation resource.
	//
	// This value will resolve to one of the following:
	// - a concrete value (e.g. `"my-awesome-bucket"`)
	// - `undefined`, when a name should be generated by CloudFormation
	// - a concrete name generated automatically during synthesis, in
	//   cross-environment scenarios.
	// Experimental.
	PhysicalName() *string
	// The stack in which this resource is defined.
	// Experimental.
	Stack() awscdk.Stack
	// The status of the application inference profile.
	//
	// ACTIVE means that the inference profile is ready to be used.
	// Experimental.
	Status() *string
	// The type of the inference profile.
	//
	// Always APPLICATION for application inference profiles.
	// Experimental.
	Type() InferenceProfileType
	// Time Stamp for Application Inference Profile update.
	// Experimental.
	UpdatedAt() *string
	// Apply the given removal policy to this resource.
	//
	// The Removal Policy controls what happens to this resource when it stops
	// being managed by CloudFormation, either because you've removed it from the
	// CDK application or because you've made a change that requires the resource
	// to be replaced.
	//
	// The resource can be deleted (`RemovalPolicy.DESTROY`), or left in your AWS
	// account for data recovery and cleanup later (`RemovalPolicy.RETAIN`).
	// Experimental.
	ApplyRemovalPolicy(policy awscdk.RemovalPolicy)
	// Experimental.
	GeneratePhysicalName() *string
	// Returns an environment-sensitive token that should be used for the resource's "ARN" attribute (e.g. `bucket.bucketArn`).
	//
	// Normally, this token will resolve to `arnAttr`, but if the resource is
	// referenced across environments, `arnComponents` will be used to synthesize
	// a concrete ARN with the resource's physical name. Make sure to reference
	// `this.physicalName` in `arnComponents`.
	// Experimental.
	GetResourceArnAttribute(arnAttr *string, arnComponents *awscdk.ArnComponents) *string
	// Returns an environment-sensitive token that should be used for the resource's "name" attribute (e.g. `bucket.bucketName`).
	//
	// Normally, this token will resolve to `nameAttr`, but if the resource is
	// referenced across environments, it will be resolved to `this.physicalName`,
	// which will be a concrete name.
	// Experimental.
	GetResourceNameAttribute(nameAttr *string) *string
	// Gives the appropriate policies to invoke and use the application inference profile.
	//
	// This method ensures the appropriate permissions are given to use either the inference profile
	// or the underlying foundation model/cross-region profile.
	// [disable-awslint:no-grants].
	//
	// Returns: An IAM Grant object representing the granted permissions.
	// Experimental.
	GrantInvoke(grantee awsiam.IGrantable) awsiam.Grant
	// Grants appropriate permissions to use the application inference profile (AIP).
	//
	// This method adds the necessary IAM permissions to allow the grantee to:
	// - Get inference profile details (bedrock:GetInferenceProfile)
	// - Invoke the model through the inference profile (bedrock:InvokeModel)
	//
	// Note: This does not grant permissions to use the underlying model/cross-region profile in the AIP.
	// For comprehensive permissions, use grantInvoke() instead.
	// [disable-awslint:no-grants].
	//
	// Returns: An IAM Grant object representing the granted permissions.
	// Experimental.
	GrantProfileUsage(grantee awsiam.IGrantable) awsiam.Grant
	// Returns a string representation of this construct.
	// Experimental.
	ToString() *string
	// Applies one or more mixins to this construct.
	//
	// Mixins are applied in order. The list of constructs is captured at the
	// start of the call, so constructs added by a mixin will not be visited.
	// Use multiple `with()` calls if subsequent mixins should apply to added
	// constructs.
	// Experimental.
	With(mixins ...constructs.IMixin) constructs.IConstruct
}

Class to create an Application Inference Profile with CDK.

These are inference profiles created by users (user defined). This helps to track costs and model usage.

Application inference profiles are user-defined profiles that help you track costs and model usage. They can be created for a single region or for multiple regions using a cross-region inference profile.

Example: 

// Create a cross-region inference profile
crossRegionProfile := bedrock.CrossRegionInferenceProfile_FromConfig(&CrossRegionInferenceProfileProps{
	GeoRegion: bedrock.CrossRegionInferenceProfileRegion_US,
	Model: bedrock.BedrockFoundationModel_ANTHROPIC_CLAUDE_3_5_SONNET_V2_0(),
})

// Create an application inference profile across regions
appProfile := bedrock.NewApplicationInferenceProfile(this, jsii.String("MyMultiRegionProfile"), &ApplicationInferenceProfileProps{
	ApplicationInferenceProfileName: jsii.String("claude-35-sonnet-v2-multi-region"),
	ModelSource: crossRegionProfile,
	Description: jsii.String("Multi-region application profile for cost tracking"),
})

See: https://docs.aws.amazon.com/bedrock/latest/userguide/inference-profiles-create.html

Experimental.

func NewApplicationInferenceProfile 

func NewApplicationInferenceProfile(scope constructs.Construct, id *string, props *ApplicationInferenceProfileProps) ApplicationInferenceProfile

Experimental.

type ApplicationInferenceProfileAttributes 

type ApplicationInferenceProfileAttributes struct {
	// The ARN of the application inference profile.
	// Experimental.
	InferenceProfileArn *string `field:"required" json:"inferenceProfileArn" yaml:"inferenceProfileArn"`
	// The ID or Amazon Resource Name (ARN) of the inference profile.
	//
	// This can be either the profile ID or the full ARN.
	// Experimental.
	InferenceProfileIdentifier *string `field:"required" json:"inferenceProfileIdentifier" yaml:"inferenceProfileIdentifier"`
}

Attributes for specifying an imported Application Inference Profile.

Example: 

// Import an inference profile through attributes
importedProfile := bedrock.ApplicationInferenceProfile_FromApplicationInferenceProfileAttributes(this, jsii.String("ImportedProfile"), &ApplicationInferenceProfileAttributes{
	InferenceProfileArn: jsii.String("arn:aws:bedrock:us-east-1:123456789012:application-inference-profile/my-profile-id"),
	InferenceProfileIdentifier: jsii.String("my-profile-id"),
})

Experimental.

type ApplicationInferenceProfileProps 

type ApplicationInferenceProfileProps struct {
	// The name of the application inference profile.
	//
	// This name will be used to identify the inference profile in the AWS console and APIs.
	// - Required:  Yes
	// - Maximum length: 64 characters
	// - Pattern: `^([0-9a-zA-Z:.][ _-]?)+$`
	// See: https://docs.aws.amazon.com/AWSCloudFormation/latest/UserGuide/aws-resource-bedrock-applicationinferenceprofile.html#cfn-bedrock-applicationinferenceprofile-inferenceprofilename
	//
	// Experimental.
	ApplicationInferenceProfileName *string `field:"required" json:"applicationInferenceProfileName" yaml:"applicationInferenceProfileName"`
	// The model source for this inference profile.
	//
	// To create an application inference profile for one Region, specify a foundation model.
	// Usage and costs for requests made to that Region with that model will be tracked.
	//
	// To create an application inference profile for multiple Regions,
	// specify a cross region (system-defined) inference profile.
	// The inference profile will route requests to the Regions defined in
	// the cross region (system-defined) inference profile that you choose.
	// Usage and costs for requests made to the Regions in the inference profile will be tracked.
	// Experimental.
	ModelSource IBedrockInvokable `field:"required" json:"modelSource" yaml:"modelSource"`
	// Description of the inference profile. Provides additional context about the purpose and usage of this inference profile.
	//
	// - Maximum length: 200 characters when provided
	// - Pattern: `^([0-9a-zA-Z:.][ _-]?)+$`
	// See: https://docs.aws.amazon.com/AWSCloudFormation/latest/UserGuide/aws-resource-bedrock-applicationinferenceprofile.html#cfn-bedrock-applicationinferenceprofile-description
	//
	// Default: - No description is provided.
	//
	// Experimental.
	Description *string `field:"optional" json:"description" yaml:"description"`
	// A list of tags associated with the inference profile.
	//
	// Tags help you organize and categorize your AWS resources.
	// Default: - No tags are applied.
	//
	// Experimental.
	Tags *map[string]*string `field:"optional" json:"tags" yaml:"tags"`
}

Properties for creating an Application Inference Profile.

Example: 

// Create a cross-region inference profile
crossRegionProfile := bedrock.CrossRegionInferenceProfile_FromConfig(&CrossRegionInferenceProfileProps{
	GeoRegion: bedrock.CrossRegionInferenceProfileRegion_US,
	Model: bedrock.BedrockFoundationModel_ANTHROPIC_CLAUDE_3_5_SONNET_V2_0(),
})

// Create an application inference profile across regions
appProfile := bedrock.NewApplicationInferenceProfile(this, jsii.String("MyMultiRegionProfile"), &ApplicationInferenceProfileProps{
	ApplicationInferenceProfileName: jsii.String("claude-35-sonnet-v2-multi-region"),
	ModelSource: crossRegionProfile,
	Description: jsii.String("Multi-region application profile for cost tracking"),
})

Experimental.

type AssetApiSchema 

type AssetApiSchema interface {
	ApiSchema
	// The inline OpenAPI schema definition as a string, if using an inline schema.
	//
	// Can be in JSON or YAML format.
	// Experimental.
	InlineSchema() *string
	// The S3 location of the API schema file, if using an S3-based schema.
	//
	// Contains the bucket name and object key information.
	// Experimental.
	S3File() *awss3.Location
	// Binds this API schema to a construct scope.
	//
	// This method initializes the S3 asset if it hasn't been initialized yet.
	// Must be called before rendering the schema as CFN properties.
	// Experimental.
	Bind(scope constructs.Construct)
}

API Schema from a local asset.

The asset is uploaded to an S3 staging bucket, then moved to its final location by CloudFormation during deployment.

Example: 

actionGroupFunction := lambda.NewFunction(this, jsii.String("ActionGroupFunction"), &FunctionProps{
	Runtime: lambda.Runtime_PYTHON_3_12(),
	Handler: jsii.String("index.handler"),
	Code: lambda.Code_FromAsset(path.join(__dirname, jsii.String("../lambda/action-group"))),
})

// When using ApiSchema.fromLocalAsset, you must bind the schema to a scope
schema := bedrock.ApiSchema_FromLocalAsset(path.join(__dirname, jsii.String("action-group.yaml")))
schema.Bind(this)

actionGroup := bedrock.NewAgentActionGroup(&AgentActionGroupProps{
	Name: jsii.String("query-library"),
	Description: jsii.String("Use these functions to get information about the books in the library."),
	Executor: bedrock.ActionGroupExecutor_FromLambda(actionGroupFunction),
	Enabled: jsii.Boolean(true),
	ApiSchema: schema,
})

agent := bedrock.NewAgent(this, jsii.String("Agent"), &AgentProps{
	FoundationModel: bedrock.BedrockFoundationModel_ANTHROPIC_CLAUDE_HAIKU_V1_0(),
	Instruction: jsii.String("You are a helpful and friendly agent that answers questions about literature."),
})

agent.AddActionGroup(actionGroup)

Experimental.

func ApiSchema_FromLocalAsset 

func ApiSchema_FromLocalAsset(path *string) AssetApiSchema

Creates an API Schema from a local file. Experimental.

func AssetApiSchema_FromLocalAsset 

func AssetApiSchema_FromLocalAsset(path *string) AssetApiSchema

Creates an API Schema from a local file. Experimental.

func InlineApiSchema_FromLocalAsset 

func InlineApiSchema_FromLocalAsset(path *string) AssetApiSchema

Creates an API Schema from a local file. Experimental.

func NewAssetApiSchema 

func NewAssetApiSchema(path *string, options *awss3assets.AssetOptions) AssetApiSchema

Experimental.

func S3ApiSchema_FromLocalAsset 

func S3ApiSchema_FromLocalAsset(path *string) AssetApiSchema

Creates an API Schema from a local file. Experimental.

type BedrockFoundationModel 

type BedrockFoundationModel interface {
	IBedrockInvokable
	// The ARN used for invoking the model.
	//
	// This is the same as modelArn for foundation models.
	// Experimental.
	InvokableArn() *string
	// The ARN of the foundation model.
	//
	// Format: arn:${Partition}:bedrock:${Region}::foundation-model/${ResourceId}.
	// Experimental.
	ModelArn() *string
	// The unique identifier of the foundation model.
	// Experimental.
	ModelId() *string
	// The vector types supported by this model for embeddings.
	//
	// Defines whether the model supports floating-point or binary vectors.
	// Experimental.
	SupportedVectorType() *[]VectorType
	// Whether this model supports integration with Bedrock Agents.
	//
	// When true, the model can be used with Bedrock Agents for automated task execution.
	// Experimental.
	SupportsAgents() *bool
	// Whether this model supports cross-region inference.
	//
	// When true, the model can be used with Cross-Region Inference Profiles.
	// Experimental.
	SupportsCrossRegion() *bool
	// Whether this model supports integration with Bedrock Knowledge Base.
	//
	// When true, the model can be used for knowledge base operations.
	// Experimental.
	SupportsKnowledgeBase() *bool
	// The dimensionality of the vector embeddings produced by this model.
	//
	// Only applicable for embedding models.
	// Experimental.
	VectorDimensions() *float64
	// Returns the ARN of the foundation model in the following format: `arn:${Partition}:bedrock:${Region}::foundation-model/${ResourceId}`.
	// Experimental.
	AsArn() *string
	// Returns the IModel.
	// Experimental.
	AsIModel() awsbedrock.IModel
	// Gives the appropriate policies to invoke and use the Foundation Model in the stack region.
	//
	// [disable-awslint:no-grants].
	// Experimental.
	GrantInvoke(grantee awsiam.IGrantable) awsiam.Grant
	// Gives the appropriate policies to invoke and use the Foundation Model in all regions.
	//
	// [disable-awslint:no-grants].
	// Experimental.
	GrantInvokeAllRegions(grantee awsiam.IGrantable) awsiam.Grant
	// Returns a string representation of an object.
	// Experimental.
	ToString() *string
}

Bedrock models.

If you need to use a model name that doesn't exist as a static member, you can instantiate a `BedrockFoundationModel` object, e.g: `new BedrockFoundationModel('my-model')`.

Example: 

parserFunction := lambda.NewFunction(this, jsii.String("ParserFunction"), &FunctionProps{
	Runtime: lambda.Runtime_PYTHON_3_10(),
	Handler: jsii.String("index.handler"),
	Code: lambda.Code_FromAsset(jsii.String("lambda")),
})

agent := bedrock.NewAgent(this, jsii.String("Agent"), &AgentProps{
	FoundationModel: bedrock.BedrockFoundationModel_AMAZON_NOVA_LITE_V1(),
	Instruction: jsii.String("You are a helpful assistant."),
	PromptOverrideConfiguration: bedrock.PromptOverrideConfiguration_WithCustomParser(&CustomParserProps{
		Parser: parserFunction,
		PreProcessingStep: &PromptPreProcessingConfigCustomParser{
			StepType: bedrock.AgentStepType_PRE_PROCESSING,
			UseCustomParser: jsii.Boolean(true),
		},
	}),
})

Experimental.

func BedrockFoundationModel_AI21_JAMBA_1_5_LARGE_V1 

func BedrockFoundationModel_AI21_JAMBA_1_5_LARGE_V1() BedrockFoundationModel

func BedrockFoundationModel_AI21_JAMBA_1_5_MINI_V1 

func BedrockFoundationModel_AI21_JAMBA_1_5_MINI_V1() BedrockFoundationModel

func BedrockFoundationModel_AI21_JAMBA_INSTRUCT_V1 

func BedrockFoundationModel_AI21_JAMBA_INSTRUCT_V1() BedrockFoundationModel

func BedrockFoundationModel_AMAZON_NOVA_LITE_V1 

func BedrockFoundationModel_AMAZON_NOVA_LITE_V1() BedrockFoundationModel

func BedrockFoundationModel_AMAZON_NOVA_MICRO_V1 

func BedrockFoundationModel_AMAZON_NOVA_MICRO_V1() BedrockFoundationModel

func BedrockFoundationModel_AMAZON_NOVA_PREMIER_V1 

func BedrockFoundationModel_AMAZON_NOVA_PREMIER_V1() BedrockFoundationModel

func BedrockFoundationModel_AMAZON_NOVA_PRO_V1 

func BedrockFoundationModel_AMAZON_NOVA_PRO_V1() BedrockFoundationModel

func BedrockFoundationModel_AMAZON_TITAN_PREMIER_V1_0 

func BedrockFoundationModel_AMAZON_TITAN_PREMIER_V1_0() BedrockFoundationModel

func BedrockFoundationModel_AMAZON_TITAN_TEXT_EXPRESS_V1 

func BedrockFoundationModel_AMAZON_TITAN_TEXT_EXPRESS_V1() BedrockFoundationModel

func BedrockFoundationModel_ANTHROPIC_CLAUDE_3_5_HAIKU_V1_0 

func BedrockFoundationModel_ANTHROPIC_CLAUDE_3_5_HAIKU_V1_0() BedrockFoundationModel

func BedrockFoundationModel_ANTHROPIC_CLAUDE_3_5_SONNET_V1_0 

func BedrockFoundationModel_ANTHROPIC_CLAUDE_3_5_SONNET_V1_0() BedrockFoundationModel

func BedrockFoundationModel_ANTHROPIC_CLAUDE_3_5_SONNET_V2_0 

func BedrockFoundationModel_ANTHROPIC_CLAUDE_3_5_SONNET_V2_0() BedrockFoundationModel

func BedrockFoundationModel_ANTHROPIC_CLAUDE_3_7_SONNET_V1_0 

func BedrockFoundationModel_ANTHROPIC_CLAUDE_3_7_SONNET_V1_0() BedrockFoundationModel

func BedrockFoundationModel_ANTHROPIC_CLAUDE_HAIKU_4_5_V1_0 

func BedrockFoundationModel_ANTHROPIC_CLAUDE_HAIKU_4_5_V1_0() BedrockFoundationModel

func BedrockFoundationModel_ANTHROPIC_CLAUDE_HAIKU_V1_0 

func BedrockFoundationModel_ANTHROPIC_CLAUDE_HAIKU_V1_0() BedrockFoundationModel

func BedrockFoundationModel_ANTHROPIC_CLAUDE_INSTANT_V1_2 

func BedrockFoundationModel_ANTHROPIC_CLAUDE_INSTANT_V1_2() BedrockFoundationModel

func BedrockFoundationModel_ANTHROPIC_CLAUDE_OPUS_4_1_V1_0 

func BedrockFoundationModel_ANTHROPIC_CLAUDE_OPUS_4_1_V1_0() BedrockFoundationModel

func BedrockFoundationModel_ANTHROPIC_CLAUDE_OPUS_4_5_V1_0 

func BedrockFoundationModel_ANTHROPIC_CLAUDE_OPUS_4_5_V1_0() BedrockFoundationModel

func BedrockFoundationModel_ANTHROPIC_CLAUDE_OPUS_4_6_V1 

func BedrockFoundationModel_ANTHROPIC_CLAUDE_OPUS_4_6_V1() BedrockFoundationModel

func BedrockFoundationModel_ANTHROPIC_CLAUDE_OPUS_4_7 

func BedrockFoundationModel_ANTHROPIC_CLAUDE_OPUS_4_7() BedrockFoundationModel

func BedrockFoundationModel_ANTHROPIC_CLAUDE_OPUS_4_V1_0 

func BedrockFoundationModel_ANTHROPIC_CLAUDE_OPUS_4_V1_0() BedrockFoundationModel

func BedrockFoundationModel_ANTHROPIC_CLAUDE_OPUS_V1_0 

func BedrockFoundationModel_ANTHROPIC_CLAUDE_OPUS_V1_0() BedrockFoundationModel

func BedrockFoundationModel_ANTHROPIC_CLAUDE_SONNET_4_5_V1_0 

func BedrockFoundationModel_ANTHROPIC_CLAUDE_SONNET_4_5_V1_0() BedrockFoundationModel

func BedrockFoundationModel_ANTHROPIC_CLAUDE_SONNET_4_6 

func BedrockFoundationModel_ANTHROPIC_CLAUDE_SONNET_4_6() BedrockFoundationModel

func BedrockFoundationModel_ANTHROPIC_CLAUDE_SONNET_4_V1_0 

func BedrockFoundationModel_ANTHROPIC_CLAUDE_SONNET_4_V1_0() BedrockFoundationModel

func BedrockFoundationModel_ANTHROPIC_CLAUDE_SONNET_V1_0 

func BedrockFoundationModel_ANTHROPIC_CLAUDE_SONNET_V1_0() BedrockFoundationModel

func BedrockFoundationModel_ANTHROPIC_CLAUDE_V2 

func BedrockFoundationModel_ANTHROPIC_CLAUDE_V2() BedrockFoundationModel

func BedrockFoundationModel_ANTHROPIC_CLAUDE_V2_1 

func BedrockFoundationModel_ANTHROPIC_CLAUDE_V2_1() BedrockFoundationModel

func BedrockFoundationModel_COHERE_EMBED_ENGLISH_V3 

func BedrockFoundationModel_COHERE_EMBED_ENGLISH_V3() BedrockFoundationModel

func BedrockFoundationModel_COHERE_EMBED_MULTILINGUAL_V3 

func BedrockFoundationModel_COHERE_EMBED_MULTILINGUAL_V3() BedrockFoundationModel

func BedrockFoundationModel_DEEPSEEK_R1_V1 

func BedrockFoundationModel_DEEPSEEK_R1_V1() BedrockFoundationModel

func BedrockFoundationModel_FromCdkFoundationModel 

func BedrockFoundationModel_FromCdkFoundationModel(modelId awsbedrock.FoundationModel, props *BedrockFoundationModelProps) BedrockFoundationModel

Creates a BedrockFoundationModel from a FoundationModel.

Use this method when you have a foundation model from the CDK.

Returns: A new BedrockFoundationModel instance. Experimental.

func BedrockFoundationModel_FromCdkFoundationModelId 

func BedrockFoundationModel_FromCdkFoundationModelId(modelId awsbedrock.FoundationModelIdentifier, props *BedrockFoundationModelProps) BedrockFoundationModel

Creates a BedrockFoundationModel from a FoundationModelIdentifier.

Use this method when you have a model identifier from the CDK.

Returns: A new BedrockFoundationModel instance. Experimental.

func BedrockFoundationModel_META_LLAMA_3_1_8B_INSTRUCT_V1 

func BedrockFoundationModel_META_LLAMA_3_1_8B_INSTRUCT_V1() BedrockFoundationModel

func BedrockFoundationModel_META_LLAMA_3_1_70B_INSTRUCT_V1 

func BedrockFoundationModel_META_LLAMA_3_1_70B_INSTRUCT_V1() BedrockFoundationModel

func BedrockFoundationModel_META_LLAMA_3_2_1B_INSTRUCT_V1 

func BedrockFoundationModel_META_LLAMA_3_2_1B_INSTRUCT_V1() BedrockFoundationModel

func BedrockFoundationModel_META_LLAMA_3_2_3B_INSTRUCT_V1 

func BedrockFoundationModel_META_LLAMA_3_2_3B_INSTRUCT_V1() BedrockFoundationModel

func BedrockFoundationModel_META_LLAMA_3_2_11B_INSTRUCT_V1 

func BedrockFoundationModel_META_LLAMA_3_2_11B_INSTRUCT_V1() BedrockFoundationModel

func BedrockFoundationModel_META_LLAMA_3_3_70B_INSTRUCT_V1 

func BedrockFoundationModel_META_LLAMA_3_3_70B_INSTRUCT_V1() BedrockFoundationModel

func BedrockFoundationModel_META_LLAMA_4_MAVERICK_17B_INSTRUCT_V1 

func BedrockFoundationModel_META_LLAMA_4_MAVERICK_17B_INSTRUCT_V1() BedrockFoundationModel

func BedrockFoundationModel_META_LLAMA_4_SCOUT_17B_INSTRUCT_V1 

func BedrockFoundationModel_META_LLAMA_4_SCOUT_17B_INSTRUCT_V1() BedrockFoundationModel

func BedrockFoundationModel_MISTRAL_7B_INSTRUCT_V0 

func BedrockFoundationModel_MISTRAL_7B_INSTRUCT_V0() BedrockFoundationModel

func BedrockFoundationModel_MISTRAL_LARGE_2402_V1 

func BedrockFoundationModel_MISTRAL_LARGE_2402_V1() BedrockFoundationModel

func BedrockFoundationModel_MISTRAL_LARGE_2407_V1 

func BedrockFoundationModel_MISTRAL_LARGE_2407_V1() BedrockFoundationModel

func BedrockFoundationModel_MISTRAL_MIXTRAL_8X7B_INSTRUCT_V0 

func BedrockFoundationModel_MISTRAL_MIXTRAL_8X7B_INSTRUCT_V0() BedrockFoundationModel

func BedrockFoundationModel_MISTRAL_PIXTRAL_LARGE_2502_V1 

func BedrockFoundationModel_MISTRAL_PIXTRAL_LARGE_2502_V1() BedrockFoundationModel

func BedrockFoundationModel_MISTRAL_SMALL_2402_V1 

func BedrockFoundationModel_MISTRAL_SMALL_2402_V1() BedrockFoundationModel

func BedrockFoundationModel_TITAN_EMBED_TEXT_V1 

func BedrockFoundationModel_TITAN_EMBED_TEXT_V1() BedrockFoundationModel

func BedrockFoundationModel_TITAN_EMBED_TEXT_V2_256 

func BedrockFoundationModel_TITAN_EMBED_TEXT_V2_256() BedrockFoundationModel

func BedrockFoundationModel_TITAN_EMBED_TEXT_V2_512 

func BedrockFoundationModel_TITAN_EMBED_TEXT_V2_512() BedrockFoundationModel

func BedrockFoundationModel_TITAN_EMBED_TEXT_V2_1024 

func BedrockFoundationModel_TITAN_EMBED_TEXT_V2_1024() BedrockFoundationModel

func NewBedrockFoundationModel 

func NewBedrockFoundationModel(value *string, props *BedrockFoundationModelProps) BedrockFoundationModel

Experimental.

type BedrockFoundationModelProps 

type BedrockFoundationModelProps struct {
	// https://docs.aws.amazon.com/bedrock/latest/userguide/model-lifecycle.html A version is marked Legacy when there is a more recent version which provides superior performance. Amazon Bedrock sets an EOL date for Legacy versions.
	// Default: - false.
	//
	// Experimental.
	Legacy *bool `field:"optional" json:"legacy" yaml:"legacy"`
	// Currently, some of the offered models are optimized with prompts/parsers fine-tuned for integrating with the agents architecture.
	//
	// When true, the model has been specifically optimized for agent interactions.
	// Default: - false.
	//
	// Experimental.
	OptimizedForAgents *bool `field:"optional" json:"optimizedForAgents" yaml:"optimizedForAgents"`
	// Embeddings models have different supported vector types.
	//
	// Defines whether the model supports floating-point or binary vectors.
	// Default: - undefined.
	//
	// Experimental.
	SupportedVectorType *[]VectorType `field:"optional" json:"supportedVectorType" yaml:"supportedVectorType"`
	// Bedrock Agents can use this model.
	//
	// When true, the model can be used with Bedrock Agents for automated task execution.
	// Default: - false.
	//
	// Experimental.
	SupportsAgents *bool `field:"optional" json:"supportsAgents" yaml:"supportsAgents"`
	// Can be used with a Cross-Region Inference Profile.
	//
	// When true, the model supports inference across different AWS regions.
	// Default: - false.
	//
	// Experimental.
	SupportsCrossRegion *bool `field:"optional" json:"supportsCrossRegion" yaml:"supportsCrossRegion"`
	// Bedrock Knowledge Base can use this model.
	//
	// When true, the model can be used for knowledge base operations.
	// Default: - false.
	//
	// Experimental.
	SupportsKnowledgeBase *bool `field:"optional" json:"supportsKnowledgeBase" yaml:"supportsKnowledgeBase"`
	// Embedding models have different vector dimensions.
	//
	// Only applicable for embedding models. Defines the dimensionality of the vector embeddings.
	// Default: - undefined.
	//
	// Experimental.
	VectorDimensions *float64 `field:"optional" json:"vectorDimensions" yaml:"vectorDimensions"`
}

Properties for configuring a Bedrock Foundation Model.

These properties define the model's capabilities and supported features.

Example: 

// The code below shows an example of how to instantiate this type.
// The values are placeholders you should change.
import bedrock_alpha "github.com/aws/aws-cdk-go/awsbedrockalpha"

bedrockFoundationModelProps := &BedrockFoundationModelProps{
	Legacy: jsii.Boolean(false),
	OptimizedForAgents: jsii.Boolean(false),
	SupportedVectorType: []VectorType{
		bedrock_alpha.VectorType_FLOATING_POINT,
	},
	SupportsAgents: jsii.Boolean(false),
	SupportsCrossRegion: jsii.Boolean(false),
	SupportsKnowledgeBase: jsii.Boolean(false),
	VectorDimensions: jsii.Number(123),
}

Experimental.

type CanadaSpecificPIIType 

type CanadaSpecificPIIType interface {
	PIIType
	// The string value of the PII type.
	// Experimental.
	Value() *string
	// Returns the string representation of the PII type.
	//
	// Returns: The string value of the PII type.
	// Experimental.
	ToString() *string
}

Types of PII specific to Canada.

Example: 

// The code below shows an example of how to instantiate this type.
// The values are placeholders you should change.
import bedrock_alpha "github.com/aws/aws-cdk-go/awsbedrockalpha"

canadaSpecificPIIType := bedrock_alpha.CanadaSpecificPIIType_CA_HEALTH_NUMBER()

Experimental.

func CanadaSpecificPIIType_CA_HEALTH_NUMBER 

func CanadaSpecificPIIType_CA_HEALTH_NUMBER() CanadaSpecificPIIType

func CanadaSpecificPIIType_CA_SOCIAL_INSURANCE_NUMBER 

func CanadaSpecificPIIType_CA_SOCIAL_INSURANCE_NUMBER() CanadaSpecificPIIType

type ChatMessage 

type ChatMessage interface {
	// The role of the message sender.
	// Experimental.
	Role() ChatMessageRole
	// The text content of the message.
	// Experimental.
	Text() *string
}

Represents a message in a chat conversation.

Example: 

cmk := kms.NewKey(this, jsii.String("cmk"), &KeyProps{
})

variantChat := bedrock.PromptVariant_Chat(&ChatPromptVariantProps{
	VariantName: jsii.String("variant1"),
	Model: bedrock.BedrockFoundationModel_ANTHROPIC_CLAUDE_3_5_SONNET_V1_0(),
	Messages: []ChatMessage{
		bedrock.ChatMessage_User(jsii.String("From now on, you speak Japanese!")),
		bedrock.ChatMessage_Assistant(jsii.String("Konnichiwa!")),
		bedrock.ChatMessage_*User(jsii.String("From now on, you speak {{language}}!")),
	},
	System: jsii.String("You are a helpful assistant that only speaks the language you`re told."),
	PromptVariables: []*string{
		jsii.String("language"),
	},
	ToolConfiguration: &ToolConfiguration{
		ToolChoice: bedrock.ToolChoice_AUTO(),
		Tools: []Tool{
			bedrock.Tool_Function(&FunctionToolProps{
				Name: jsii.String("top_song"),
				Description: jsii.String("Get the most popular song played on a radio station."),
				InputSchema: map[string]interface{}{
					"type": jsii.String("object"),
					"properties": map[string]map[string]*string{
						"sign": map[string]*string{
							"type": jsii.String("string"),
							"description": jsii.String("The call sign for the radio station for which you want the most popular song. Example calls signs are WZPZ and WKR."),
						},
					},
					"required": []*string{
						jsii.String("sign"),
					},
				},
			}),
		},
	},
})

bedrock.NewPrompt(this, jsii.String("prompt1"), &PromptProps{
	PromptName: jsii.String("prompt-chat"),
	Description: jsii.String("my first chat prompt"),
	DefaultVariant: variantChat,
	Variants: []IPromptVariant{
		variantChat,
	},
	KmsKey: cmk,
})

Experimental.

func ChatMessage_Assistant 

func ChatMessage_Assistant(text *string) ChatMessage

Creates an assistant message.

Returns: A ChatMessage instance representing an assistant message. Experimental.

func ChatMessage_User 

func ChatMessage_User(text *string) ChatMessage

Creates a user message.

Returns: A ChatMessage instance representing a user message. Experimental.

func NewChatMessage 

func NewChatMessage(role ChatMessageRole, text *string) ChatMessage

Experimental.

type ChatMessageRole 

type ChatMessageRole string

The role of a message in a chat conversation. Experimental. 

const (
	// This role represents the human user in the conversation.
	//
	// Inputs from the
	// user guide the conversation and prompt responses from the assistant.
	// Experimental.
	ChatMessageRole_USER ChatMessageRole = "USER"
	// This is the role of the model itself, responding to user inputs based on the context set by the system.
	// Experimental.
	ChatMessageRole_ASSISTANT ChatMessageRole = "ASSISTANT"
)

type ChatPromptVariantProps 

type ChatPromptVariantProps struct {
	// The model which is used to run the prompt.
	//
	// The model could be a foundation
	// model, a custom model, or a provisioned model.
	// Experimental.
	Model IBedrockInvokable `field:"required" json:"model" yaml:"model"`
	// The name of the prompt variant.
	// Experimental.
	VariantName *string `field:"required" json:"variantName" yaml:"variantName"`
	// The variables in the prompt template that can be filled in at runtime.
	// Default: - No variables defined.
	//
	// Experimental.
	PromptVariables *[]*string `field:"optional" json:"promptVariables" yaml:"promptVariables"`
	// The messages in the chat prompt.
	//
	// Must include at least one User Message.
	// The messages should alternate between User and Assistant.
	// Experimental.
	Messages *[]ChatMessage `field:"required" json:"messages" yaml:"messages"`
	// Inference configuration for the Chat Prompt.
	// Default: - No inference configuration provided.
	//
	// Experimental.
	InferenceConfiguration PromptInferenceConfiguration `field:"optional" json:"inferenceConfiguration" yaml:"inferenceConfiguration"`
	// Context or instructions for the model to consider before generating a response.
	// Default: - No system message provided.
	//
	// Experimental.
	System *string `field:"optional" json:"system" yaml:"system"`
	// The configuration with available tools to the model and how it must use them.
	// Default: - No tool configuration provided.
	//
	// Experimental.
	ToolConfiguration *ToolConfiguration `field:"optional" json:"toolConfiguration" yaml:"toolConfiguration"`
}

Properties for creating a chat prompt variant.

Example: 

cmk := kms.NewKey(this, jsii.String("cmk"), &KeyProps{
})

variantChat := bedrock.PromptVariant_Chat(&ChatPromptVariantProps{
	VariantName: jsii.String("variant1"),
	Model: bedrock.BedrockFoundationModel_ANTHROPIC_CLAUDE_3_5_SONNET_V1_0(),
	Messages: []ChatMessage{
		bedrock.ChatMessage_User(jsii.String("From now on, you speak Japanese!")),
		bedrock.ChatMessage_Assistant(jsii.String("Konnichiwa!")),
		bedrock.ChatMessage_*User(jsii.String("From now on, you speak {{language}}!")),
	},
	System: jsii.String("You are a helpful assistant that only speaks the language you`re told."),
	PromptVariables: []*string{
		jsii.String("language"),
	},
	ToolConfiguration: &ToolConfiguration{
		ToolChoice: bedrock.ToolChoice_AUTO(),
		Tools: []Tool{
			bedrock.Tool_Function(&FunctionToolProps{
				Name: jsii.String("top_song"),
				Description: jsii.String("Get the most popular song played on a radio station."),
				InputSchema: map[string]interface{}{
					"type": jsii.String("object"),
					"properties": map[string]map[string]*string{
						"sign": map[string]*string{
							"type": jsii.String("string"),
							"description": jsii.String("The call sign for the radio station for which you want the most popular song. Example calls signs are WZPZ and WKR."),
						},
					},
					"required": []*string{
						jsii.String("sign"),
					},
				},
			}),
		},
	},
})

bedrock.NewPrompt(this, jsii.String("prompt1"), &PromptProps{
	PromptName: jsii.String("prompt-chat"),
	Description: jsii.String("my first chat prompt"),
	DefaultVariant: variantChat,
	Variants: []IPromptVariant{
		variantChat,
	},
	KmsKey: cmk,
})

Experimental.

type ChatTemplateConfigurationProps 

type ChatTemplateConfigurationProps struct {
	// The messages in the chat template.
	// Experimental.
	Messages *[]ChatMessage `field:"required" json:"messages" yaml:"messages"`
	// The input variables for the template.
	// Default: - No input variables.
	//
	// Experimental.
	InputVariables *[]*string `field:"optional" json:"inputVariables" yaml:"inputVariables"`
	// The system message for the chat template.
	// Default: - No system message.
	//
	// Experimental.
	System *string `field:"optional" json:"system" yaml:"system"`
	// The tool configuration for the chat template.
	// Default: - No tool configuration.
	//
	// Experimental.
	ToolConfiguration *ToolConfiguration `field:"optional" json:"toolConfiguration" yaml:"toolConfiguration"`
}

Properties for creating a chat template configuration.

Example: 

// The code below shows an example of how to instantiate this type.
// The values are placeholders you should change.
import bedrock_alpha "github.com/aws/aws-cdk-go/awsbedrockalpha"

var chatMessage ChatMessage
var tool Tool
var toolChoice ToolChoice

chatTemplateConfigurationProps := &ChatTemplateConfigurationProps{
	Messages: []ChatMessage{
		chatMessage,
	},

	// the properties below are optional
	InputVariables: []*string{
		jsii.String("inputVariables"),
	},
	System: jsii.String("system"),
	ToolConfiguration: &ToolConfiguration{
		ToolChoice: toolChoice,
		Tools: []Tool{
			tool,
		},
	},
}

Experimental.

type CommonPromptVariantProps 

type CommonPromptVariantProps struct {
	// The model which is used to run the prompt.
	//
	// The model could be a foundation
	// model, a custom model, or a provisioned model.
	// Experimental.
	Model IBedrockInvokable `field:"required" json:"model" yaml:"model"`
	// The name of the prompt variant.
	// Experimental.
	VariantName *string `field:"required" json:"variantName" yaml:"variantName"`
	// The variables in the prompt template that can be filled in at runtime.
	// Default: - No variables defined.
	//
	// Experimental.
	PromptVariables *[]*string `field:"optional" json:"promptVariables" yaml:"promptVariables"`
}

Common properties for all prompt variants.

Example: 

// The code below shows an example of how to instantiate this type.
// The values are placeholders you should change.
import bedrock_alpha "github.com/aws/aws-cdk-go/awsbedrockalpha"

var bedrockInvokable IBedrockInvokable

commonPromptVariantProps := &CommonPromptVariantProps{
	Model: bedrockInvokable,
	VariantName: jsii.String("variantName"),

	// the properties below are optional
	PromptVariables: []*string{
		jsii.String("promptVariables"),
	},
}

Experimental.

type ContentFilter 

type ContentFilter struct {
	// The strength of the content filter to apply to prompts / user input.
	// Experimental.
	InputStrength ContentFilterStrength `field:"required" json:"inputStrength" yaml:"inputStrength"`
	// The strength of the content filter to apply to model responses.
	// Experimental.
	OutputStrength ContentFilterStrength `field:"required" json:"outputStrength" yaml:"outputStrength"`
	// The type of harmful category that the content filter is applied to.
	// Experimental.
	Type ContentFilterType `field:"required" json:"type" yaml:"type"`
	// The action to take when content is detected in the input.
	// Default: GuardrailAction.BLOCK
	//
	// Experimental.
	InputAction GuardrailAction `field:"optional" json:"inputAction" yaml:"inputAction"`
	// Whether the content filter is enabled for input.
	// Default: true.
	//
	// Experimental.
	InputEnabled *bool `field:"optional" json:"inputEnabled" yaml:"inputEnabled"`
	// The input modalities to apply the content filter to.
	// Default: undefined - Applies to text modality.
	//
	// Experimental.
	InputModalities *[]ModalityType `field:"optional" json:"inputModalities" yaml:"inputModalities"`
	// The action to take when content is detected in the output.
	// Default: GuardrailAction.BLOCK
	//
	// Experimental.
	OutputAction GuardrailAction `field:"optional" json:"outputAction" yaml:"outputAction"`
	// Whether the content filter is enabled for output.
	// Default: true.
	//
	// Experimental.
	OutputEnabled *bool `field:"optional" json:"outputEnabled" yaml:"outputEnabled"`
	// The output modalities to apply the content filter to.
	// Default: undefined - Applies to text modality.
	//
	// Experimental.
	OutputModalities *[]ModalityType `field:"optional" json:"outputModalities" yaml:"outputModalities"`
}

Interface to declare a content filter.

Example: 

// Create a guardrail to filter inappropriate content
guardrail := bedrock.NewGuardrail(this, jsii.String("bedrockGuardrails"), &GuardrailProps{
	GuardrailName: jsii.String("my-BedrockGuardrails"),
	Description: jsii.String("Legal ethical guardrails."),
})

guardrail.AddContentFilter(&ContentFilter{
	Type: bedrock.ContentFilterType_SEXUAL,
	InputStrength: bedrock.ContentFilterStrength_HIGH,
	OutputStrength: bedrock.ContentFilterStrength_MEDIUM,
})

// Create an agent with the guardrail
agentWithGuardrail := bedrock.NewAgent(this, jsii.String("AgentWithGuardrail"), &AgentProps{
	FoundationModel: bedrock.BedrockFoundationModel_ANTHROPIC_CLAUDE_HAIKU_V1_0(),
	Instruction: jsii.String("You are a helpful and friendly agent that answers questions about literature."),
	Guardrail: guardrail,
})

Experimental.

type ContentFilterStrength 

type ContentFilterStrength string

The strength of the content filter.

As you increase the filter strength, the likelihood of filtering harmful content increases and the probability of seeing harmful content in your application reduces.

Example: 

// Create a guardrail to filter inappropriate content
guardrail := bedrock.NewGuardrail(this, jsii.String("bedrockGuardrails"), &GuardrailProps{
	GuardrailName: jsii.String("my-BedrockGuardrails"),
	Description: jsii.String("Legal ethical guardrails."),
})

guardrail.AddContentFilter(&ContentFilter{
	Type: bedrock.ContentFilterType_SEXUAL,
	InputStrength: bedrock.ContentFilterStrength_HIGH,
	OutputStrength: bedrock.ContentFilterStrength_MEDIUM,
})

// Create an agent with the guardrail
agentWithGuardrail := bedrock.NewAgent(this, jsii.String("AgentWithGuardrail"), &AgentProps{
	FoundationModel: bedrock.BedrockFoundationModel_ANTHROPIC_CLAUDE_HAIKU_V1_0(),
	Instruction: jsii.String("You are a helpful and friendly agent that answers questions about literature."),
	Guardrail: guardrail,
})

Experimental. 

const (
	// No content filtering applied.
	// Experimental.
	ContentFilterStrength_NONE ContentFilterStrength = "NONE"
	// Low strength content filtering - minimal filtering of harmful content.
	// Experimental.
	ContentFilterStrength_LOW ContentFilterStrength = "LOW"
	// Medium strength content filtering - balanced filtering of harmful content.
	// Experimental.
	ContentFilterStrength_MEDIUM ContentFilterStrength = "MEDIUM"
	// High strength content filtering - aggressive filtering of harmful content.
	// Experimental.
	ContentFilterStrength_HIGH ContentFilterStrength = "HIGH"
)

type ContentFilterType 

type ContentFilterType string

The type of harmful category usable in a content filter.

Example: 

// Create a guardrail to filter inappropriate content
guardrail := bedrock.NewGuardrail(this, jsii.String("bedrockGuardrails"), &GuardrailProps{
	GuardrailName: jsii.String("my-BedrockGuardrails"),
	Description: jsii.String("Legal ethical guardrails."),
})

guardrail.AddContentFilter(&ContentFilter{
	Type: bedrock.ContentFilterType_SEXUAL,
	InputStrength: bedrock.ContentFilterStrength_HIGH,
	OutputStrength: bedrock.ContentFilterStrength_MEDIUM,
})

// Create an agent with the guardrail
agentWithGuardrail := bedrock.NewAgent(this, jsii.String("AgentWithGuardrail"), &AgentProps{
	FoundationModel: bedrock.BedrockFoundationModel_ANTHROPIC_CLAUDE_HAIKU_V1_0(),
	Instruction: jsii.String("You are a helpful and friendly agent that answers questions about literature."),
	Guardrail: guardrail,
})

Experimental. 

const (
	// Describes input prompts and model responses that indicates sexual interest, activity, or arousal using direct or indirect references to body parts, physical traits, or sex.
	// Experimental.
	ContentFilterType_SEXUAL ContentFilterType = "SEXUAL"
	// Describes input prompts and model responses that includes glorification of or threats to inflict physical pain, hurt, or injury toward a person, group or thing.
	// Experimental.
	ContentFilterType_VIOLENCE ContentFilterType = "VIOLENCE"
	// Describes input prompts and model responses that discriminate, criticize, insult, denounce, or dehumanize a person or group on the basis of an identity (such as race, ethnicity, gender, religion, sexual orientation, ability, and national origin).
	// Experimental.
	ContentFilterType_HATE ContentFilterType = "HATE"
	// Describes input prompts and model responses that includes demeaning, humiliating, mocking, insulting, or belittling language.
	//
	// This type of language is also labeled
	// as bullying.
	// Experimental.
	ContentFilterType_INSULTS ContentFilterType = "INSULTS"
	// Describes input prompts and model responses that seeks or provides information about engaging in misconduct activity, or harming, defrauding, or taking advantage of a person, group or institution.
	// Experimental.
	ContentFilterType_MISCONDUCT ContentFilterType = "MISCONDUCT"
	// Enable to detect and block user inputs attempting to override system instructions.
	//
	// To avoid misclassifying system prompts as a prompt attack and ensure that the filters
	// are selectively applied to user inputs, use input tagging.
	// Experimental.
	ContentFilterType_PROMPT_ATTACK ContentFilterType = "PROMPT_ATTACK"
)

type ContextualGroundingFilter 

type ContextualGroundingFilter struct {
	// The threshold for the contextual grounding filter.
	//
	// - `0` (blocks nothing)
	// - `0.99` (blocks almost everything)
	// Experimental.
	Threshold *float64 `field:"required" json:"threshold" yaml:"threshold"`
	// The type of contextual grounding filter.
	// Experimental.
	Type ContextualGroundingFilterType `field:"required" json:"type" yaml:"type"`
	// The action to take when contextual grounding is detected.
	// Default: GuardrailAction.BLOCK
	//
	// Experimental.
	Action GuardrailAction `field:"optional" json:"action" yaml:"action"`
	// Whether the contextual grounding filter is enabled.
	// Default: true.
	//
	// Experimental.
	Enabled *bool `field:"optional" json:"enabled" yaml:"enabled"`
}

Interface to define a Contextual Grounding Filter.

Example: 

guardrail := bedrock.NewGuardrail(this, jsii.String("bedrockGuardrails"), &GuardrailProps{
	GuardrailName: jsii.String("my-BedrockGuardrails"),
})
// Add contextual grounding filter with action and enabled flag
guardrail.AddContextualGroundingFilter(&ContextualGroundingFilter{
	Type: bedrock.ContextualGroundingFilterType_GROUNDING,
	Threshold: jsii.Number(0.8),
	// the properties below are optional
	Action: bedrock.GuardrailAction_BLOCK,
	Enabled: jsii.Boolean(true),
})

Experimental.

type ContextualGroundingFilterType 

type ContextualGroundingFilterType string

The type of contextual grounding filter.

Example: 

guardrail := bedrock.NewGuardrail(this, jsii.String("bedrockGuardrails"), &GuardrailProps{
	GuardrailName: jsii.String("my-BedrockGuardrails"),
})
// Add contextual grounding filter with action and enabled flag
guardrail.AddContextualGroundingFilter(&ContextualGroundingFilter{
	Type: bedrock.ContextualGroundingFilterType_GROUNDING,
	Threshold: jsii.Number(0.8),
	// the properties below are optional
	Action: bedrock.GuardrailAction_BLOCK,
	Enabled: jsii.Boolean(true),
})

Experimental. 

const (
	// Grounding score represents the confidence that the model response is factually correct and grounded in the source.
	//
	// If the model response has a lower score than
	// the defined threshold, the response will be blocked and the configured blocked
	// message will be returned to the user. A higher threshold level blocks more responses.
	// Experimental.
	ContextualGroundingFilterType_GROUNDING ContextualGroundingFilterType = "GROUNDING"
	// Relevance score represents the confidence that the model response is relevant to the user's query.
	//
	// If the model response has a lower score than the defined
	// threshold, the response will be blocked and the configured blocked message will
	// be returned to the user. A higher threshold level blocks more responses.
	// Experimental.
	ContextualGroundingFilterType_RELEVANCE ContextualGroundingFilterType = "RELEVANCE"
)

type CrossRegionInferenceProfile 

type CrossRegionInferenceProfile interface {
	IBedrockInvokable
	IInferenceProfile
	// The ARN of the inference profile.
	// Experimental.
	InferenceProfileArn() *string
	// The unique identifier of the inference profile.
	//
	// Format: {geoRegion}.{modelId}
	// Experimental.
	InferenceProfileId() *string
	// The underlying foundation model supporting cross-region inference.
	// Experimental.
	InferenceProfileModel() BedrockFoundationModel
	// The ARN used for invoking this inference profile.
	//
	// This equals to the inferenceProfileArn property, useful for implementing IBedrockInvokable interface.
	// Experimental.
	InvokableArn() *string
	// The type of inference profile.
	//
	// Always SYSTEM_DEFINED for cross-region profiles.
	// Experimental.
	Type() InferenceProfileType
	// Gives the appropriate policies to invoke and use the Foundation Model.
	//
	// For cross-region inference profiles, this method grants permissions to:
	// - Invoke the model in all regions where the inference profile can route requests
	// - Use the inference profile itself
	// [disable-awslint:no-grants].
	//
	// Returns: An IAM Grant object representing the granted permissions.
	// Experimental.
	GrantInvoke(grantee awsiam.IGrantable) awsiam.Grant
	// Grants appropriate permissions to use the cross-region inference profile.
	//
	// This method adds the necessary IAM permissions to allow the grantee to:
	// - Get inference profile details (bedrock:GetInferenceProfile)
	// - Invoke the model through the inference profile (bedrock:InvokeModel*)
	//
	// Note: This does not grant permissions to use the underlying model directly.
	// For comprehensive permissions, use grantInvoke() instead.
	// [disable-awslint:no-grants].
	//
	// Returns: An IAM Grant object representing the granted permissions.
	// Experimental.
	GrantProfileUsage(grantee awsiam.IGrantable) awsiam.Grant
}

Cross-region inference enables you to seamlessly manage unplanned traffic bursts by utilizing compute across different AWS Regions.

With cross-region inference, you can distribute traffic across multiple AWS Regions, enabling higher throughput and enhanced resilience during periods of peak demands.

This construct represents a system-defined inference profile that routes requests across multiple regions based on availability and demand.

Example: 

// Create a cross-region inference profile
crossRegionProfile := bedrock.CrossRegionInferenceProfile_FromConfig(&CrossRegionInferenceProfileProps{
	GeoRegion: bedrock.CrossRegionInferenceProfileRegion_US,
	Model: bedrock.BedrockFoundationModel_ANTHROPIC_CLAUDE_3_5_SONNET_V1_0(),
})

// Use the cross-region profile with an agent
agent := bedrock.NewAgent(this, jsii.String("Agent"), &AgentProps{
	FoundationModel: crossRegionProfile,
	Instruction: jsii.String("You are a helpful and friendly agent that answers questions about agriculture."),
})

See: https://docs.aws.amazon.com/bedrock/latest/userguide/cross-region-inference.html

Experimental.

func CrossRegionInferenceProfile_FromConfig 

func CrossRegionInferenceProfile_FromConfig(config *CrossRegionInferenceProfileProps) CrossRegionInferenceProfile

Creates a Cross-Region Inference Profile from the provided configuration.

Returns: A new CrossRegionInferenceProfile instance. Experimental.

type CrossRegionInferenceProfileProps 

type CrossRegionInferenceProfileProps struct {
	// The geographic region where the traffic is going to be distributed.
	//
	// Routing
	// factors in user traffic, demand and utilization of resources.
	// Experimental.
	GeoRegion CrossRegionInferenceProfileRegion `field:"required" json:"geoRegion" yaml:"geoRegion"`
	// A foundation model supporting cross-region inference.
	//
	// The model must have cross-region support enabled.
	// See: https://docs.aws.amazon.com/bedrock/latest/userguide/cross-region-inference-support.html
	//
	// Experimental.
	Model BedrockFoundationModel `field:"required" json:"model" yaml:"model"`
}

Properties for creating a Cross-Region Inference Profile.

Example: 

// Create a cross-region inference profile
crossRegionProfile := bedrock.CrossRegionInferenceProfile_FromConfig(&CrossRegionInferenceProfileProps{
	GeoRegion: bedrock.CrossRegionInferenceProfileRegion_US,
	Model: bedrock.BedrockFoundationModel_ANTHROPIC_CLAUDE_3_5_SONNET_V1_0(),
})

// Use the cross-region profile with an agent
agent := bedrock.NewAgent(this, jsii.String("Agent"), &AgentProps{
	FoundationModel: crossRegionProfile,
	Instruction: jsii.String("You are a helpful and friendly agent that answers questions about agriculture."),
})

Experimental.

type CrossRegionInferenceProfileRegion 

type CrossRegionInferenceProfileRegion string

Geographic regions supported for cross-region inference profiles.

These regions help distribute traffic across multiple AWS regions for better throughput and resilience during peak demands.

Example: 

// Create a cross-region inference profile
crossRegionProfile := bedrock.CrossRegionInferenceProfile_FromConfig(&CrossRegionInferenceProfileProps{
	GeoRegion: bedrock.CrossRegionInferenceProfileRegion_US,
	Model: bedrock.BedrockFoundationModel_ANTHROPIC_CLAUDE_3_5_SONNET_V1_0(),
})

// Use the cross-region profile with an agent
agent := bedrock.NewAgent(this, jsii.String("Agent"), &AgentProps{
	FoundationModel: crossRegionProfile,
	Instruction: jsii.String("You are a helpful and friendly agent that answers questions about agriculture."),
})

Experimental. 

const (
	// Global cross-region Inference Identifier.
	//
	// Routes requests to any supported commercial AWS Region.
	// Experimental.
	CrossRegionInferenceProfileRegion_GLOBAL CrossRegionInferenceProfileRegion = "GLOBAL"
	// Cross-region Inference Identifier for the European area.
	//
	// According to the model chosen, this might include:
	// - Frankfurt (`eu-central-1`)
	// - Ireland (`eu-west-1`)
	// - Paris (`eu-west-3`)
	// - London (`eu-west-2`)
	// - Stockholm (`eu-north-1`)
	// - Milan (`eu-south-1`)
	// - Spain (`eu-south-2`)
	// - Zurich (`eu-central-2`).
	// Experimental.
	CrossRegionInferenceProfileRegion_EU CrossRegionInferenceProfileRegion = "EU"
	// Cross-region Inference Identifier for the United States area.
	//
	// According to the model chosen, this might include:
	// - N. Virginia (`us-east-1`)
	// - Ohio (`us-east-2`)
	// - Oregon (`us-west-2`).
	// Experimental.
	CrossRegionInferenceProfileRegion_US CrossRegionInferenceProfileRegion = "US"
	// Cross-region Inference Identifier for the US GovCloud area.
	//
	// According to the model chosen, this might include:
	// - GovCloud US-East (`us-gov-east-1`)
	// - GovCloud US-West (`us-gov-west-1`).
	// Experimental.
	CrossRegionInferenceProfileRegion_US_GOV CrossRegionInferenceProfileRegion = "US_GOV"
	// Cross-region Inference Identifier for the Asia-Pacific area.
	//
	// According to the model chosen, this might include:
	// - Tokyo (`ap-northeast-1`)
	// - Seoul (`ap-northeast-2`)
	// - Osaka (`ap-northeast-3`)
	// - Mumbai (`ap-south-1`)
	// - Hyderabad (`ap-south-2`)
	// - Singapore (`ap-southeast-1`)
	// - Sydney (`ap-southeast-2`)
	// - Jakarta (`ap-southeast-3`)
	// - Melbourne (`ap-southeast-4`)
	// - Malaysia (`ap-southeast-5`)
	// - Thailand (`ap-southeast-7`)
	// - Taipei (`ap-east-2`)
	// - Middle East (UAE) (`me-central-1`).
	// Experimental.
	CrossRegionInferenceProfileRegion_APAC CrossRegionInferenceProfileRegion = "APAC"
	// Cross-region Inference Identifier for the Japan area.
	//
	// According to the model chosen, this might include:
	// - Tokyo (`ap-northeast-1`)
	// - Osaka (`ap-northeast-3`).
	// Experimental.
	CrossRegionInferenceProfileRegion_JP CrossRegionInferenceProfileRegion = "JP"
	// Cross-region Inference Identifier for the Australia area.
	//
	// According to the model chosen, this might include:
	// - Sydney (`ap-southeast-2`)
	// - Melbourne (`ap-southeast-4`).
	// Experimental.
	CrossRegionInferenceProfileRegion_AU CrossRegionInferenceProfileRegion = "AU"
)

type CustomControl 

type CustomControl string

The type of custom control for the action group executor. Experimental. 

const (
	// Returns the action group invocation results directly in the InvokeAgent response.
	// Experimental.
	CustomControl_RETURN_CONTROL CustomControl = "RETURN_CONTROL"
)

type CustomOrchestrationExecutor 

type CustomOrchestrationExecutor interface {
	// The Lambda function that contains the custom orchestration logic.
	//
	// This function is called when the agent needs to make decisions about action execution.
	// Experimental.
	LambdaFunction() awslambda.IFunction
	// The type of orchestration this executor performs.
	// Experimental.
	Type() OrchestrationType
}

Contains details about the Lambda function containing the orchestration logic carried out upon invoking the custom orchestration.

Example: 

orchestrationFunction := lambda.NewFunction(this, jsii.String("OrchestrationFunction"), &FunctionProps{
	Runtime: lambda.Runtime_PYTHON_3_10(),
	Handler: jsii.String("index.handler"),
	Code: lambda.Code_FromAsset(jsii.String("lambda/orchestration")),
})

agent := bedrock.NewAgent(this, jsii.String("CustomOrchestrationAgent"), &AgentProps{
	Instruction: jsii.String("You are a helpful assistant with custom orchestration logic."),
	FoundationModel: bedrock.BedrockFoundationModel_AMAZON_NOVA_LITE_V1(),
	CustomOrchestrationExecutor: bedrock.CustomOrchestrationExecutor_FromLambda(orchestrationFunction),
})

Experimental.

func CustomOrchestrationExecutor_FromLambda 

func CustomOrchestrationExecutor_FromLambda(lambdaFunction awslambda.IFunction) CustomOrchestrationExecutor

Defines an orchestration executor with a Lambda function containing the business logic. Experimental.

type CustomParserProps 

type CustomParserProps struct {
	// Configuration for the knowledge base response generation step.
	// Default: undefined - No knowledge base response generation configuration.
	//
	// Experimental.
	KnowledgeBaseResponseGenerationStep *PromptKnowledgeBaseResponseGenerationConfigCustomParser `field:"optional" json:"knowledgeBaseResponseGenerationStep" yaml:"knowledgeBaseResponseGenerationStep"`
	// Configuration for the memory summarization step.
	// Default: undefined - No memory summarization configuration.
	//
	// Experimental.
	MemorySummarizationStep *PromptMemorySummarizationConfigCustomParser `field:"optional" json:"memorySummarizationStep" yaml:"memorySummarizationStep"`
	// Configuration for the orchestration step.
	// Default: undefined - No orchestration configuration.
	//
	// Experimental.
	OrchestrationStep *PromptOrchestrationConfigCustomParser `field:"optional" json:"orchestrationStep" yaml:"orchestrationStep"`
	// Lambda function to use as custom parser.
	// Default: undefined - No custom parser is used.
	//
	// Experimental.
	Parser awslambda.IFunction `field:"optional" json:"parser" yaml:"parser"`
	// Configuration for the post-processing step.
	// Default: undefined - No post-processing configuration.
	//
	// Experimental.
	PostProcessingStep *PromptPostProcessingConfigCustomParser `field:"optional" json:"postProcessingStep" yaml:"postProcessingStep"`
	// Configuration for the pre-processing step.
	// Default: undefined - No pre-processing configuration.
	//
	// Experimental.
	PreProcessingStep *PromptPreProcessingConfigCustomParser `field:"optional" json:"preProcessingStep" yaml:"preProcessingStep"`
	// Configuration for the routing classifier step.
	// Default: undefined - No routing classifier configuration.
	//
	// Experimental.
	RoutingClassifierStep *PromptRoutingClassifierConfigCustomParser `field:"optional" json:"routingClassifierStep" yaml:"routingClassifierStep"`
}

Properties for configuring a custom Lambda parser for prompt overrides.

Example: 

parserFunction := lambda.NewFunction(this, jsii.String("ParserFunction"), &FunctionProps{
	Runtime: lambda.Runtime_PYTHON_3_10(),
	Handler: jsii.String("index.handler"),
	Code: lambda.Code_FromAsset(jsii.String("lambda")),
})

agent := bedrock.NewAgent(this, jsii.String("Agent"), &AgentProps{
	FoundationModel: bedrock.BedrockFoundationModel_AMAZON_NOVA_LITE_V1(),
	Instruction: jsii.String("You are a helpful assistant."),
	PromptOverrideConfiguration: bedrock.PromptOverrideConfiguration_WithCustomParser(&CustomParserProps{
		Parser: parserFunction,
		PreProcessingStep: &PromptPreProcessingConfigCustomParser{
			StepType: bedrock.AgentStepType_PRE_PROCESSING,
			UseCustomParser: jsii.Boolean(true),
		},
	}),
})

Experimental.

type CustomTopicProps 

type CustomTopicProps struct {
	// Provide a clear definition to detect and block user inputs and FM responses that fall into this topic.
	//
	// Avoid starting with "don't".
	//
	// Example:
	//   `Investment advice refers to inquiries, guidance, or recommendations
	//   regarding the management or allocation of funds or assets with the goal of
	//   generating returns or achieving specific financial objectives.`
	//
	// Experimental.
	Definition *string `field:"required" json:"definition" yaml:"definition"`
	// Representative phrases that refer to the topic.
	//
	// These phrases can represent
	// a user input or a model response. Add between 1 and 100 phrases, up to 100 characters
	// each.
	//
	// Example:
	//   "Where should I invest my money?"
	//
	// Experimental.
	Examples *[]*string `field:"required" json:"examples" yaml:"examples"`
	// The name of the topic to deny.
	// Experimental.
	Name *string `field:"required" json:"name" yaml:"name"`
	// The action to take when a topic is detected in the input.
	// Default: GuardrailAction.BLOCK
	//
	// Experimental.
	InputAction GuardrailAction `field:"optional" json:"inputAction" yaml:"inputAction"`
	// Whether the topic filter is enabled for input.
	// Default: true.
	//
	// Experimental.
	InputEnabled *bool `field:"optional" json:"inputEnabled" yaml:"inputEnabled"`
	// The action to take when a topic is detected in the output.
	// Default: GuardrailAction.BLOCK
	//
	// Experimental.
	OutputAction GuardrailAction `field:"optional" json:"outputAction" yaml:"outputAction"`
	// Whether the topic filter is enabled for output.
	// Default: true.
	//
	// Experimental.
	OutputEnabled *bool `field:"optional" json:"outputEnabled" yaml:"outputEnabled"`
}

Interface for creating a custom Topic.

Example: 

guardrail := bedrock.NewGuardrail(this, jsii.String("bedrockGuardrails"), &GuardrailProps{
	GuardrailName: jsii.String("my-BedrockGuardrails"),
	// Configure tier for topic filters (optional)
	TopicsTierConfig: bedrock.TierConfig_STANDARD,
})

// Use a predefined topic
guardrail.AddDeniedTopicFilter(bedrock.Topic_FINANCIAL_ADVICE())

// Create a custom topic with input/output actions
guardrail.AddDeniedTopicFilter(bedrock.Topic_Custom(&CustomTopicProps{
	Name: jsii.String("Legal_Advice"),
	Definition: jsii.String("Offering guidance or suggestions on legal matters, legal actions, interpretation of laws, or legal rights and responsibilities."),
	Examples: []*string{
		jsii.String("Can I sue someone for this?"),
		jsii.String("What are my legal rights in this situation?"),
		jsii.String("Is this action against the law?"),
		jsii.String("What should I do to file a legal complaint?"),
		jsii.String("Can you explain this law to me?"),
	},
	// props below are optional
	InputAction: bedrock.GuardrailAction_BLOCK,
	InputEnabled: jsii.Boolean(true),
	OutputAction: bedrock.GuardrailAction_NONE,
	OutputEnabled: jsii.Boolean(true),
}))

Experimental.

type DefaultPromptRouterIdentifier 

type DefaultPromptRouterIdentifier interface {
	// The unique identifier for this prompt router.
	// Experimental.
	PromptRouterId() *string
	// The foundation models that this router can route between.
	// Experimental.
	RoutingModels() *[]BedrockFoundationModel
}

Represents identifiers for default prompt routers in Bedrock.

These are pre-configured routers provided by AWS that route between different models in the same family for optimal performance and cost.

Example: 

// Create a prompt router for intelligent model selection
promptRouter := bedrock.PromptRouter_FromDefaultId(bedrock.DefaultPromptRouterIdentifier_ANTHROPIC_CLAUDE_V1(), jsii.String("us-east-1"))

// Use the prompt router with a prompt variant
variant := bedrock.PromptVariant_Text(&TextPromptVariantProps{
	VariantName: jsii.String("variant1"),
	PromptText: jsii.String("What is the capital of France?"),
	Model: promptRouter,
})

bedrock.NewPrompt(this, jsii.String("Prompt"), &PromptProps{
	PromptName: jsii.String("prompt-router-test"),
	Variants: []IPromptVariant{
		variant,
	},
})

Experimental.

func DefaultPromptRouterIdentifier_ANTHROPIC_CLAUDE_V1 

func DefaultPromptRouterIdentifier_ANTHROPIC_CLAUDE_V1() DefaultPromptRouterIdentifier

func DefaultPromptRouterIdentifier_META_LLAMA_3_1 

func DefaultPromptRouterIdentifier_META_LLAMA_3_1() DefaultPromptRouterIdentifier

type FinancePIIType 

type FinancePIIType interface {
	PIIType
	// The string value of the PII type.
	// Experimental.
	Value() *string
	// Returns the string representation of the PII type.
	//
	// Returns: The string value of the PII type.
	// Experimental.
	ToString() *string
}

Types of PII in the domain of Finance.

Example: 

guardrail := bedrock.NewGuardrail(this, jsii.String("bedrockGuardrails"), &GuardrailProps{
	GuardrailName: jsii.String("my-BedrockGuardrails"),
})

// Add PII filter for addresses with input/output actions
guardrail.AddPIIFilter(&PIIFilter{
	Type: bedrock.GeneralPIIType_ADDRESS(),
	Action: bedrock.GuardrailAction_BLOCK,
	// below props are optional
	InputAction: bedrock.GuardrailAction_BLOCK,
	InputEnabled: jsii.Boolean(true),
	OutputAction: bedrock.GuardrailAction_ANONYMIZE,
	OutputEnabled: jsii.Boolean(true),
})

// Add PII filter for credit card numbers with input/output actions
guardrail.AddPIIFilter(&PIIFilter{
	Type: bedrock.FinancePIIType_CREDIT_DEBIT_CARD_NUMBER(),
	Action: bedrock.GuardrailAction_BLOCK,
	// below props are optional
	InputAction: bedrock.GuardrailAction_BLOCK,
	InputEnabled: jsii.Boolean(true),
	OutputAction: bedrock.GuardrailAction_ANONYMIZE,
	OutputEnabled: jsii.Boolean(true),
})

// Add PII filter for email addresses
guardrail.AddPIIFilter(&PIIFilter{
	Type: bedrock.GeneralPIIType_EMAIL(),
	Action: bedrock.GuardrailAction_ANONYMIZE,
})

// Add PII filter for US Social Security Numbers
guardrail.AddPIIFilter(&PIIFilter{
	Type: bedrock.USASpecificPIIType_US_SOCIAL_SECURITY_NUMBER(),
	Action: bedrock.GuardrailAction_BLOCK,
})

// Add PII filter for IP addresses
guardrail.AddPIIFilter(&PIIFilter{
	Type: bedrock.InformationTechnologyPIIType_IP_ADDRESS(),
	Action: bedrock.GuardrailAction_ANONYMIZE,
})

Experimental.

func FinancePIIType_CREDIT_DEBIT_CARD_CVV 

func FinancePIIType_CREDIT_DEBIT_CARD_CVV() FinancePIIType

func FinancePIIType_CREDIT_DEBIT_CARD_EXPIRY 

func FinancePIIType_CREDIT_DEBIT_CARD_EXPIRY() FinancePIIType

func FinancePIIType_CREDIT_DEBIT_CARD_NUMBER 

func FinancePIIType_CREDIT_DEBIT_CARD_NUMBER() FinancePIIType

func FinancePIIType_INTERNATIONAL_BANK_ACCOUNT_NUMBER 

func FinancePIIType_INTERNATIONAL_BANK_ACCOUNT_NUMBER() FinancePIIType

func FinancePIIType_PIN 

func FinancePIIType_PIN() FinancePIIType

func FinancePIIType_SWIFT_CODE 

func FinancePIIType_SWIFT_CODE() FinancePIIType

type Function 

type Function interface {
	// Description of the function.
	// Experimental.
	Description() *string
	// The name of the function.
	// Experimental.
	Name() *string
	// Parameters for the function.
	// Experimental.
	Parameters() *map[string]FunctionParameter
	// Whether to require confirmation before executing the function.
	// Experimental.
	RequireConfirmation() RequireConfirmation
}

Represents a function in a function schema.

Example: 

// The code below shows an example of how to instantiate this type.
// The values are placeholders you should change.
import "github.com/aws/aws-cdk-go/awsbedrockalpha"

function_ := bedrock_alpha.NewFunction(&FunctionProps{
	Description: jsii.String("description"),
	Name: jsii.String("name"),

	// the properties below are optional
	Parameters: map[string]FunctionParameterProps{
		"parametersKey": &FunctionParameterProps{
			"type": bedrock_alpha.ParameterType_STRING,

			// the properties below are optional
			"description": jsii.String("description"),
			"required": jsii.Boolean(false),
		},
	},
	RequireConfirmation: bedrock_alpha.RequireConfirmation_ENABLED,
})

Experimental.

func NewFunction 

func NewFunction(props *FunctionProps) Function

Experimental.

type FunctionParameter 

type FunctionParameter interface {
	// Description of the parameter.
	// Default: undefined no description will be present.
	//
	// Experimental.
	Description() *string
	// Whether the parameter is required.
	// Experimental.
	Required() *bool
	// The type of the parameter.
	// Experimental.
	Type() ParameterType
}

Represents a function parameter in a function schema.

Example: 

// The code below shows an example of how to instantiate this type.
// The values are placeholders you should change.
import "github.com/aws/aws-cdk-go/awsbedrockalpha"

functionParameter := bedrock_alpha.NewFunctionParameter(&FunctionParameterProps{
	Type: bedrock_alpha.ParameterType_STRING,

	// the properties below are optional
	Description: jsii.String("description"),
	Required: jsii.Boolean(false),
})

Experimental.

func NewFunctionParameter 

func NewFunctionParameter(props *FunctionParameterProps) FunctionParameter

Experimental.

type FunctionParameterProps 

type FunctionParameterProps struct {
	// The type of the parameter.
	// Experimental.
	Type ParameterType `field:"required" json:"type" yaml:"type"`
	// Description of the parameter.
	// Default: undefined no description will be present.
	//
	// Experimental.
	Description *string `field:"optional" json:"description" yaml:"description"`
	// Whether the parameter is required.
	// Default: true.
	//
	// Experimental.
	Required *bool `field:"optional" json:"required" yaml:"required"`
}

Properties for a function parameter.

Example: 

actionGroupFunction := lambda.NewFunction(this, jsii.String("ActionGroupFunction"), &FunctionProps{
	Runtime: lambda.Runtime_PYTHON_3_12(),
	Handler: jsii.String("index.handler"),
	Code: lambda.Code_FromAsset(path.join(__dirname, jsii.String("../lambda/action-group"))),
})

// Define a function schema with parameters
functionSchema := bedrock.NewFunctionSchema(&FunctionSchemaProps{
	Functions: []FunctionProps{
		&FunctionProps{
			Name: jsii.String("searchBooks"),
			Description: jsii.String("Search for books in the library catalog"),
			Parameters: map[string]FunctionParameterProps{
				"query": &FunctionParameterProps{
					"type": bedrock.ParameterType_STRING,
					"required": jsii.Boolean(true),
					"description": jsii.String("The search query string"),
				},
				"maxResults": &FunctionParameterProps{
					"type": bedrock.ParameterType_INTEGER,
					"required": jsii.Boolean(false),
					"description": jsii.String("Maximum number of results to return"),
				},
				"includeOutOfPrint": &FunctionParameterProps{
					"type": bedrock.ParameterType_BOOLEAN,
					"required": jsii.Boolean(false),
					"description": jsii.String("Whether to include out-of-print books"),
				},
			},
			RequireConfirmation: bedrock.RequireConfirmation_DISABLED,
		},
		&FunctionProps{
			Name: jsii.String("getBookDetails"),
			Description: jsii.String("Get detailed information about a specific book"),
			Parameters: map[string]FunctionParameterProps{
				"bookId": &FunctionParameterProps{
					"type": bedrock.ParameterType_STRING,
					"required": jsii.Boolean(true),
					"description": jsii.String("The unique identifier of the book"),
				},
			},
			RequireConfirmation: bedrock.RequireConfirmation_ENABLED,
		},
	},
})

// Create an action group using the function schema
actionGroup := bedrock.NewAgentActionGroup(&AgentActionGroupProps{
	Name: jsii.String("library-functions"),
	Description: jsii.String("Functions for interacting with the library catalog"),
	Executor: bedrock.ActionGroupExecutor_FromLambda(actionGroupFunction),
	FunctionSchema: functionSchema,
	Enabled: jsii.Boolean(true),
})

agent := bedrock.NewAgent(this, jsii.String("Agent"), &AgentProps{
	FoundationModel: bedrock.BedrockFoundationModel_ANTHROPIC_CLAUDE_HAIKU_V1_0(),
	Instruction: jsii.String("You are a helpful and friendly agent that answers questions about literature."),
	ActionGroups: []AgentActionGroup{
		actionGroup,
	},
})

Experimental.

type FunctionProps 

type FunctionProps struct {
	// Description of the function.
	// Experimental.
	Description *string `field:"required" json:"description" yaml:"description"`
	// The name of the function.
	// Experimental.
	Name *string `field:"required" json:"name" yaml:"name"`
	// Parameters for the function as a record of parameter name to parameter properties.
	// Default: {}.
	//
	// Experimental.
	Parameters *map[string]*FunctionParameterProps `field:"optional" json:"parameters" yaml:"parameters"`
	// Whether to require confirmation before executing the function.
	// Default: RequireConfirmation.DISABLED
	//
	// Experimental.
	RequireConfirmation RequireConfirmation `field:"optional" json:"requireConfirmation" yaml:"requireConfirmation"`
}

Properties for a function in a function schema.

Example: 

// The code below shows an example of how to instantiate this type.
// The values are placeholders you should change.
import "github.com/aws/aws-cdk-go/awsbedrockalpha"

functionProps := &FunctionProps{
	Description: jsii.String("description"),
	Name: jsii.String("name"),

	// the properties below are optional
	Parameters: map[string]FunctionParameterProps{
		"parametersKey": &FunctionParameterProps{
			"type": bedrock_alpha.ParameterType_STRING,

			// the properties below are optional
			"description": jsii.String("description"),
			"required": jsii.Boolean(false),
		},
	},
	RequireConfirmation: bedrock_alpha.RequireConfirmation_ENABLED,
}

Experimental.

type FunctionSchema 

type FunctionSchema interface {
	// The functions defined in the schema.
	// Experimental.
	Functions() *[]Function
}

Represents a function schema for a Bedrock Agent Action Group.

Example: 

actionGroupFunction := lambda.NewFunction(this, jsii.String("ActionGroupFunction"), &FunctionProps{
	Runtime: lambda.Runtime_PYTHON_3_12(),
	Handler: jsii.String("index.handler"),
	Code: lambda.Code_FromAsset(path.join(__dirname, jsii.String("../lambda/action-group"))),
})

// Define a function schema with parameters
functionSchema := bedrock.NewFunctionSchema(&FunctionSchemaProps{
	Functions: []FunctionProps{
		&FunctionProps{
			Name: jsii.String("searchBooks"),
			Description: jsii.String("Search for books in the library catalog"),
			Parameters: map[string]FunctionParameterProps{
				"query": &FunctionParameterProps{
					"type": bedrock.ParameterType_STRING,
					"required": jsii.Boolean(true),
					"description": jsii.String("The search query string"),
				},
				"maxResults": &FunctionParameterProps{
					"type": bedrock.ParameterType_INTEGER,
					"required": jsii.Boolean(false),
					"description": jsii.String("Maximum number of results to return"),
				},
				"includeOutOfPrint": &FunctionParameterProps{
					"type": bedrock.ParameterType_BOOLEAN,
					"required": jsii.Boolean(false),
					"description": jsii.String("Whether to include out-of-print books"),
				},
			},
			RequireConfirmation: bedrock.RequireConfirmation_DISABLED,
		},
		&FunctionProps{
			Name: jsii.String("getBookDetails"),
			Description: jsii.String("Get detailed information about a specific book"),
			Parameters: map[string]FunctionParameterProps{
				"bookId": &FunctionParameterProps{
					"type": bedrock.ParameterType_STRING,
					"required": jsii.Boolean(true),
					"description": jsii.String("The unique identifier of the book"),
				},
			},
			RequireConfirmation: bedrock.RequireConfirmation_ENABLED,
		},
	},
})

// Create an action group using the function schema
actionGroup := bedrock.NewAgentActionGroup(&AgentActionGroupProps{
	Name: jsii.String("library-functions"),
	Description: jsii.String("Functions for interacting with the library catalog"),
	Executor: bedrock.ActionGroupExecutor_FromLambda(actionGroupFunction),
	FunctionSchema: functionSchema,
	Enabled: jsii.Boolean(true),
})

agent := bedrock.NewAgent(this, jsii.String("Agent"), &AgentProps{
	FoundationModel: bedrock.BedrockFoundationModel_ANTHROPIC_CLAUDE_HAIKU_V1_0(),
	Instruction: jsii.String("You are a helpful and friendly agent that answers questions about literature."),
	ActionGroups: []AgentActionGroup{
		actionGroup,
	},
})

Experimental.

func NewFunctionSchema 

func NewFunctionSchema(props *FunctionSchemaProps) FunctionSchema

Experimental.

type FunctionSchemaProps 

type FunctionSchemaProps struct {
	// Functions defined in the schema.
	// Experimental.
	Functions *[]*FunctionProps `field:"required" json:"functions" yaml:"functions"`
}

Properties for a function schema.

Example: 

actionGroupFunction := lambda.NewFunction(this, jsii.String("ActionGroupFunction"), &FunctionProps{
	Runtime: lambda.Runtime_PYTHON_3_12(),
	Handler: jsii.String("index.handler"),
	Code: lambda.Code_FromAsset(path.join(__dirname, jsii.String("../lambda/action-group"))),
})

// Define a function schema with parameters
functionSchema := bedrock.NewFunctionSchema(&FunctionSchemaProps{
	Functions: []FunctionProps{
		&FunctionProps{
			Name: jsii.String("searchBooks"),
			Description: jsii.String("Search for books in the library catalog"),
			Parameters: map[string]FunctionParameterProps{
				"query": &FunctionParameterProps{
					"type": bedrock.ParameterType_STRING,
					"required": jsii.Boolean(true),
					"description": jsii.String("The search query string"),
				},
				"maxResults": &FunctionParameterProps{
					"type": bedrock.ParameterType_INTEGER,
					"required": jsii.Boolean(false),
					"description": jsii.String("Maximum number of results to return"),
				},
				"includeOutOfPrint": &FunctionParameterProps{
					"type": bedrock.ParameterType_BOOLEAN,
					"required": jsii.Boolean(false),
					"description": jsii.String("Whether to include out-of-print books"),
				},
			},
			RequireConfirmation: bedrock.RequireConfirmation_DISABLED,
		},
		&FunctionProps{
			Name: jsii.String("getBookDetails"),
			Description: jsii.String("Get detailed information about a specific book"),
			Parameters: map[string]FunctionParameterProps{
				"bookId": &FunctionParameterProps{
					"type": bedrock.ParameterType_STRING,
					"required": jsii.Boolean(true),
					"description": jsii.String("The unique identifier of the book"),
				},
			},
			RequireConfirmation: bedrock.RequireConfirmation_ENABLED,
		},
	},
})

// Create an action group using the function schema
actionGroup := bedrock.NewAgentActionGroup(&AgentActionGroupProps{
	Name: jsii.String("library-functions"),
	Description: jsii.String("Functions for interacting with the library catalog"),
	Executor: bedrock.ActionGroupExecutor_FromLambda(actionGroupFunction),
	FunctionSchema: functionSchema,
	Enabled: jsii.Boolean(true),
})

agent := bedrock.NewAgent(this, jsii.String("Agent"), &AgentProps{
	FoundationModel: bedrock.BedrockFoundationModel_ANTHROPIC_CLAUDE_HAIKU_V1_0(),
	Instruction: jsii.String("You are a helpful and friendly agent that answers questions about literature."),
	ActionGroups: []AgentActionGroup{
		actionGroup,
	},
})

Experimental.

type FunctionToolProps 

type FunctionToolProps struct {
	// A description of what the function does.
	// Experimental.
	Description *string `field:"required" json:"description" yaml:"description"`
	// The input schema for the function parameters.
	// Experimental.
	InputSchema interface{} `field:"required" json:"inputSchema" yaml:"inputSchema"`
	// The name of the function.
	// Experimental.
	Name *string `field:"required" json:"name" yaml:"name"`
}

Properties for creating a function tool.

Example: 

cmk := kms.NewKey(this, jsii.String("cmk"), &KeyProps{
})

variantChat := bedrock.PromptVariant_Chat(&ChatPromptVariantProps{
	VariantName: jsii.String("variant1"),
	Model: bedrock.BedrockFoundationModel_ANTHROPIC_CLAUDE_3_5_SONNET_V1_0(),
	Messages: []ChatMessage{
		bedrock.ChatMessage_User(jsii.String("From now on, you speak Japanese!")),
		bedrock.ChatMessage_Assistant(jsii.String("Konnichiwa!")),
		bedrock.ChatMessage_*User(jsii.String("From now on, you speak {{language}}!")),
	},
	System: jsii.String("You are a helpful assistant that only speaks the language you`re told."),
	PromptVariables: []*string{
		jsii.String("language"),
	},
	ToolConfiguration: &ToolConfiguration{
		ToolChoice: bedrock.ToolChoice_AUTO(),
		Tools: []Tool{
			bedrock.Tool_Function(&FunctionToolProps{
				Name: jsii.String("top_song"),
				Description: jsii.String("Get the most popular song played on a radio station."),
				InputSchema: map[string]interface{}{
					"type": jsii.String("object"),
					"properties": map[string]map[string]*string{
						"sign": map[string]*string{
							"type": jsii.String("string"),
							"description": jsii.String("The call sign for the radio station for which you want the most popular song. Example calls signs are WZPZ and WKR."),
						},
					},
					"required": []*string{
						jsii.String("sign"),
					},
				},
			}),
		},
	},
})

bedrock.NewPrompt(this, jsii.String("prompt1"), &PromptProps{
	PromptName: jsii.String("prompt-chat"),
	Description: jsii.String("my first chat prompt"),
	DefaultVariant: variantChat,
	Variants: []IPromptVariant{
		variantChat,
	},
	KmsKey: cmk,
})

Experimental.

type GeneralPIIType 

type GeneralPIIType interface {
	PIIType
	// The string value of the PII type.
	// Experimental.
	Value() *string
	// Returns the string representation of the PII type.
	//
	// Returns: The string value of the PII type.
	// Experimental.
	ToString() *string
}

Types of PII that are general, and not domain-specific.

Example: 

guardrail := bedrock.NewGuardrail(this, jsii.String("bedrockGuardrails"), &GuardrailProps{
	GuardrailName: jsii.String("my-BedrockGuardrails"),
})

// Add PII filter for addresses with input/output actions
guardrail.AddPIIFilter(&PIIFilter{
	Type: bedrock.GeneralPIIType_ADDRESS(),
	Action: bedrock.GuardrailAction_BLOCK,
	// below props are optional
	InputAction: bedrock.GuardrailAction_BLOCK,
	InputEnabled: jsii.Boolean(true),
	OutputAction: bedrock.GuardrailAction_ANONYMIZE,
	OutputEnabled: jsii.Boolean(true),
})

// Add PII filter for credit card numbers with input/output actions
guardrail.AddPIIFilter(&PIIFilter{
	Type: bedrock.FinancePIIType_CREDIT_DEBIT_CARD_NUMBER(),
	Action: bedrock.GuardrailAction_BLOCK,
	// below props are optional
	InputAction: bedrock.GuardrailAction_BLOCK,
	InputEnabled: jsii.Boolean(true),
	OutputAction: bedrock.GuardrailAction_ANONYMIZE,
	OutputEnabled: jsii.Boolean(true),
})

// Add PII filter for email addresses
guardrail.AddPIIFilter(&PIIFilter{
	Type: bedrock.GeneralPIIType_EMAIL(),
	Action: bedrock.GuardrailAction_ANONYMIZE,
})

// Add PII filter for US Social Security Numbers
guardrail.AddPIIFilter(&PIIFilter{
	Type: bedrock.USASpecificPIIType_US_SOCIAL_SECURITY_NUMBER(),
	Action: bedrock.GuardrailAction_BLOCK,
})

// Add PII filter for IP addresses
guardrail.AddPIIFilter(&PIIFilter{
	Type: bedrock.InformationTechnologyPIIType_IP_ADDRESS(),
	Action: bedrock.GuardrailAction_ANONYMIZE,
})

Experimental.

func GeneralPIIType_ADDRESS 

func GeneralPIIType_ADDRESS() GeneralPIIType

func GeneralPIIType_AGE 

func GeneralPIIType_AGE() GeneralPIIType

func GeneralPIIType_DRIVER_ID 

func GeneralPIIType_DRIVER_ID() GeneralPIIType

func GeneralPIIType_EMAIL 

func GeneralPIIType_EMAIL() GeneralPIIType

func GeneralPIIType_LICENSE_PLATE 

func GeneralPIIType_LICENSE_PLATE() GeneralPIIType

func GeneralPIIType_NAME 

func GeneralPIIType_NAME() GeneralPIIType

func GeneralPIIType_PASSWORD 

func GeneralPIIType_PASSWORD() GeneralPIIType

func GeneralPIIType_PHONE 

func GeneralPIIType_PHONE() GeneralPIIType

func GeneralPIIType_USERNAME 

func GeneralPIIType_USERNAME() GeneralPIIType

func GeneralPIIType_VEHICLE_IDENTIFICATION_NUMBER 

func GeneralPIIType_VEHICLE_IDENTIFICATION_NUMBER() GeneralPIIType

type Guardrail 

type Guardrail interface {
	GuardrailBase
	// The message to return when the guardrail blocks a prompt.
	// Default: "Sorry, your query violates our usage policy."
	//
	// Experimental.
	BlockedInputMessaging() *string
	// The message to return when the guardrail blocks a model response.
	// Default: "Sorry, I am unable to answer your question because of our usage policy."
	//
	// Experimental.
	BlockedOutputsMessaging() *string
	// The content filters applied by the guardrail.
	// Experimental.
	ContentFilters() *[]*ContentFilter
	// The tier that your guardrail uses for content filters.
	//
	// Consider using a tier that balances performance, accuracy, and compatibility with your existing generative AI workflows.
	// Default: filters.TierConfig.CLASSIC
	//
	// Experimental.
	ContentFiltersTierConfig() TierConfig
	// The contextual grounding filters applied by the guardrail.
	// Experimental.
	ContextualGroundingFilters() *[]*ContextualGroundingFilter
	// The cross-region configuration for the guardrail.
	// Experimental.
	CrossRegionConfig() *GuardrailCrossRegionConfigProperty
	// The denied topic filters applied by the guardrail.
	// Experimental.
	DeniedTopics() *[]Topic
	// The environment this resource belongs to.
	//
	// For resources that are created and managed in a Stack (those created by
	// creating new class instances like `new Role()`, `new Bucket()`, etc.), this
	// is always the same as the environment of the stack they belong to.
	//
	// For referenced resources (those obtained from referencing methods like
	// `Role.fromRoleArn()`, `Bucket.fromBucketName()`, etc.), they might be
	// different than the stack they were imported into.
	// Experimental.
	Env() *interfaces.ResourceEnvironment
	// The ARN of the guardrail.
	// Experimental.
	GuardrailArn() *string
	// The ID of the guardrail.
	// Experimental.
	GuardrailId() *string
	// The version of the guardrail.
	// Experimental.
	GuardrailVersion() *string
	// The computed hash of the guardrail properties.
	// Experimental.
	Hash() *string
	// The KMS key used to encrypt data.
	// Default: undefined - "Data is encrypted by default with a key that AWS owns and manages for you".
	//
	// Experimental.
	KmsKey() awskms.IKey
	// When this guardrail was last updated.
	// Experimental.
	LastUpdated() *string
	// The managed word list filters applied by the guardrail.
	// Experimental.
	ManagedWordListFilters() *[]*ManagedWordFilter
	// The name of the guardrail.
	// Experimental.
	Name() *string
	// The tree node.
	// Experimental.
	Node() constructs.Node
	// Returns a string-encoded token that resolves to the physical name that should be passed to the CloudFormation resource.
	//
	// This value will resolve to one of the following:
	// - a concrete value (e.g. `"my-awesome-bucket"`)
	// - `undefined`, when a name should be generated by CloudFormation
	// - a concrete name generated automatically during synthesis, in
	//   cross-environment scenarios.
	// Experimental.
	PhysicalName() *string
	// The PII filters applied by the guardrail.
	// Experimental.
	PiiFilters() *[]*PIIFilter
	// The regex filters applied by the guardrail.
	// Experimental.
	RegexFilters() *[]*RegexFilter
	// The stack in which this resource is defined.
	// Experimental.
	Stack() awscdk.Stack
	// The tier that your guardrail uses for denied topic filters.
	// Default: filters.TierConfig.CLASSIC
	//
	// Experimental.
	TopicsTierConfig() TierConfig
	// The word filters applied by the guardrail.
	// Experimental.
	WordFilters() *[]*WordFilter
	// Adds a content filter to the guardrail.
	// Experimental.
	AddContentFilter(filter *ContentFilter)
	// Adds a contextual grounding filter to the guardrail.
	// Experimental.
	AddContextualGroundingFilter(filter *ContextualGroundingFilter)
	// Adds a denied topic filter to the guardrail.
	// Experimental.
	AddDeniedTopicFilter(filter Topic)
	// Adds a managed word list filter to the guardrail.
	// Experimental.
	AddManagedWordListFilter(filter *ManagedWordFilter)
	// Adds a PII filter to the guardrail.
	// Experimental.
	AddPIIFilter(filter *PIIFilter)
	// Adds a regex filter to the guardrail.
	// Experimental.
	AddRegexFilter(filter *RegexFilter)
	// Adds a word filter to the guardrail.
	// Experimental.
	AddWordFilter(filter *WordFilter)
	// Adds a word filter to the guardrail.
	// Experimental.
	AddWordFilterFromFile(filePath *string, inputAction GuardrailAction, outputAction GuardrailAction, inputEnabled *bool, outputEnabled *bool)
	// Apply the given removal policy to this resource.
	//
	// The Removal Policy controls what happens to this resource when it stops
	// being managed by CloudFormation, either because you've removed it from the
	// CDK application or because you've made a change that requires the resource
	// to be replaced.
	//
	// The resource can be deleted (`RemovalPolicy.DESTROY`), or left in your AWS
	// account for data recovery and cleanup later (`RemovalPolicy.RETAIN`).
	// Experimental.
	ApplyRemovalPolicy(policy awscdk.RemovalPolicy)
	// Create a version for the guardrail.
	//
	// Returns: The guardrail version.
	// Experimental.
	CreateVersion(description *string) *string
	// Experimental.
	GeneratePhysicalName() *string
	// Returns an environment-sensitive token that should be used for the resource's "ARN" attribute (e.g. `bucket.bucketArn`).
	//
	// Normally, this token will resolve to `arnAttr`, but if the resource is
	// referenced across environments, `arnComponents` will be used to synthesize
	// a concrete ARN with the resource's physical name. Make sure to reference
	// `this.physicalName` in `arnComponents`.
	// Experimental.
	GetResourceArnAttribute(arnAttr *string, arnComponents *awscdk.ArnComponents) *string
	// Returns an environment-sensitive token that should be used for the resource's "name" attribute (e.g. `bucket.bucketName`).
	//
	// Normally, this token will resolve to `nameAttr`, but if the resource is
	// referenced across environments, it will be resolved to `this.physicalName`,
	// which will be a concrete name.
	// Experimental.
	GetResourceNameAttribute(nameAttr *string) *string
	// Grant the given principal identity permissions to perform actions on this guardrail.
	//
	// [disable-awslint:no-grants].
	// Experimental.
	Grant(grantee awsiam.IGrantable, actions ...*string) awsiam.Grant
	// Grant the given identity permissions to apply the guardrail.
	//
	// [disable-awslint:no-grants].
	// Experimental.
	GrantApply(grantee awsiam.IGrantable) awsiam.Grant
	// Return the given named metric for this guardrail.
	//
	// By default, the metric will be calculated as a sum over a period of 5 minutes.
	// You can customize this by using the `statistic` and `period` properties.
	// Experimental.
	Metric(metricName *string, props *awscloudwatch.MetricOptions) awscloudwatch.Metric
	// Return the invocation client errors metric for this guardrail.
	// Experimental.
	MetricInvocationClientErrors(props *awscloudwatch.MetricOptions) awscloudwatch.Metric
	// Return the invocation latency metric for this guardrail.
	// Experimental.
	MetricInvocationLatency(props *awscloudwatch.MetricOptions) awscloudwatch.Metric
	// Return the invocations metric for this guardrail.
	// Experimental.
	MetricInvocations(props *awscloudwatch.MetricOptions) awscloudwatch.Metric
	// Return the invocation server errors metric for this guardrail.
	// Experimental.
	MetricInvocationServerErrors(props *awscloudwatch.MetricOptions) awscloudwatch.Metric
	// Return the invocations intervened metric for this guardrail.
	// Experimental.
	MetricInvocationsIntervened(props *awscloudwatch.MetricOptions) awscloudwatch.Metric
	// Return the invocation throttles metric for this guardrail.
	// Experimental.
	MetricInvocationThrottles(props *awscloudwatch.MetricOptions) awscloudwatch.Metric
	// Return the text unit count metric for this guardrail.
	// Experimental.
	MetricTextUnitCount(props *awscloudwatch.MetricOptions) awscloudwatch.Metric
	// Returns a string representation of this construct.
	// Experimental.
	ToString() *string
	// Experimental.
	UpdateVersion(version *string)
	// Applies one or more mixins to this construct.
	//
	// Mixins are applied in order. The list of constructs is captured at the
	// start of the call, so constructs added by a mixin will not be visited.
	// Use multiple `with()` calls if subsequent mixins should apply to added
	// constructs.
	// Experimental.
	With(mixins ...constructs.IMixin) constructs.IConstruct
}

Class to create a Guardrail with CDK.

Example: 

guardrail := bedrock.NewGuardrail(this, jsii.String("bedrockGuardrails"), &GuardrailProps{
	GuardrailName: jsii.String("my-BedrockGuardrails"),
})
// Add regex filter with input/output actions
guardrail.AddRegexFilter(&RegexFilter{
	Name: jsii.String("TestRegexFilter"),
	Pattern: jsii.String("test-pattern"),
	Action: bedrock.GuardrailAction_ANONYMIZE,
	// below props are optional
	Description: jsii.String("This is a test regex filter"),
	InputAction: bedrock.GuardrailAction_BLOCK,
	InputEnabled: jsii.Boolean(true),
	OutputAction: bedrock.GuardrailAction_ANONYMIZE,
	OutputEnabled: jsii.Boolean(true),
})

Experimental.

func NewGuardrail 

func NewGuardrail(scope constructs.Construct, id *string, props *GuardrailProps) Guardrail

Experimental.

type GuardrailAction 

type GuardrailAction string

Guardrail action when a sensitive entity is detected.

Example: 

guardrail := bedrock.NewGuardrail(this, jsii.String("bedrockGuardrails"), &GuardrailProps{
	GuardrailName: jsii.String("my-BedrockGuardrails"),
	// Configure tier for topic filters (optional)
	TopicsTierConfig: bedrock.TierConfig_STANDARD,
})

// Use a predefined topic
guardrail.AddDeniedTopicFilter(bedrock.Topic_FINANCIAL_ADVICE())

// Create a custom topic with input/output actions
guardrail.AddDeniedTopicFilter(bedrock.Topic_Custom(&CustomTopicProps{
	Name: jsii.String("Legal_Advice"),
	Definition: jsii.String("Offering guidance or suggestions on legal matters, legal actions, interpretation of laws, or legal rights and responsibilities."),
	Examples: []*string{
		jsii.String("Can I sue someone for this?"),
		jsii.String("What are my legal rights in this situation?"),
		jsii.String("Is this action against the law?"),
		jsii.String("What should I do to file a legal complaint?"),
		jsii.String("Can you explain this law to me?"),
	},
	// props below are optional
	InputAction: bedrock.GuardrailAction_BLOCK,
	InputEnabled: jsii.Boolean(true),
	OutputAction: bedrock.GuardrailAction_NONE,
	OutputEnabled: jsii.Boolean(true),
}))

Experimental. 

const (
	// If sensitive information is detected in the prompt or response, the guardrail blocks all the content and returns a message that you configure.
	// Experimental.
	GuardrailAction_BLOCK GuardrailAction = "BLOCK"
	// If sensitive information is detected in the model response, the guardrail masks it with an identifier, the sensitive information is masked and replaced with identifier tags (for example: [NAME-1], [NAME-2], [EMAIL-1], etc.).
	// Experimental.
	GuardrailAction_ANONYMIZE GuardrailAction = "ANONYMIZE"
	// Do not take any action.
	// Experimental.
	GuardrailAction_NONE GuardrailAction = "NONE"
)

type GuardrailAttributes 

type GuardrailAttributes struct {
	// The ARN of the guardrail.
	//
	// At least one of guardrailArn or guardrailId must be
	// defined in order to initialize a guardrail ref.
	// Experimental.
	GuardrailArn *string `field:"required" json:"guardrailArn" yaml:"guardrailArn"`
	// The version of the guardrail.
	// Default: "DRAFT".
	//
	// Experimental.
	GuardrailVersion *string `field:"optional" json:"guardrailVersion" yaml:"guardrailVersion"`
	// The KMS key of the guardrail if custom encryption is configured.
	// Default: undefined - Means data is encrypted by default with a AWS-managed key.
	//
	// Experimental.
	KmsKey awskms.IKey `field:"optional" json:"kmsKey" yaml:"kmsKey"`
}

**************************************************************************** ATTRS FOR IMPORTED CONSTRUCT ***************************************************************************.

Example: 

var stack Stack

cmk := kms.NewKey(this, jsii.String("cmk"), &KeyProps{
})
// Import an existing guardrail by ARN
importedGuardrail := bedrock.Guardrail_FromGuardrailAttributes(stack, jsii.String("TestGuardrail"), &GuardrailAttributes{
	GuardrailArn: jsii.String("arn:aws:bedrock:us-east-1:123456789012:guardrail/oygh3o8g7rtl"),
	GuardrailVersion: jsii.String("1"),
	 //optional
	KmsKey: cmk,
})

Experimental.

type GuardrailBase 

type GuardrailBase interface {
	awscdk.Resource
	IGuardrail
	// The environment this resource belongs to.
	//
	// For resources that are created and managed in a Stack (those created by
	// creating new class instances like `new Role()`, `new Bucket()`, etc.), this
	// is always the same as the environment of the stack they belong to.
	//
	// For referenced resources (those obtained from referencing methods like
	// `Role.fromRoleArn()`, `Bucket.fromBucketName()`, etc.), they might be
	// different than the stack they were imported into.
	// Experimental.
	Env() *interfaces.ResourceEnvironment
	// The ARN of the guardrail.
	// Experimental.
	GuardrailArn() *string
	// The ID of the guardrail.
	// Experimental.
	GuardrailId() *string
	// The version of the guardrail.
	// Experimental.
	GuardrailVersion() *string
	// Optional KMS encryption key associated with this guardrail.
	// Experimental.
	KmsKey() awskms.IKey
	// When this guardrail was last updated.
	// Experimental.
	LastUpdated() *string
	// The tree node.
	// Experimental.
	Node() constructs.Node
	// Returns a string-encoded token that resolves to the physical name that should be passed to the CloudFormation resource.
	//
	// This value will resolve to one of the following:
	// - a concrete value (e.g. `"my-awesome-bucket"`)
	// - `undefined`, when a name should be generated by CloudFormation
	// - a concrete name generated automatically during synthesis, in
	//   cross-environment scenarios.
	// Experimental.
	PhysicalName() *string
	// The stack in which this resource is defined.
	// Experimental.
	Stack() awscdk.Stack
	// Apply the given removal policy to this resource.
	//
	// The Removal Policy controls what happens to this resource when it stops
	// being managed by CloudFormation, either because you've removed it from the
	// CDK application or because you've made a change that requires the resource
	// to be replaced.
	//
	// The resource can be deleted (`RemovalPolicy.DESTROY`), or left in your AWS
	// account for data recovery and cleanup later (`RemovalPolicy.RETAIN`).
	// Experimental.
	ApplyRemovalPolicy(policy awscdk.RemovalPolicy)
	// Experimental.
	GeneratePhysicalName() *string
	// Returns an environment-sensitive token that should be used for the resource's "ARN" attribute (e.g. `bucket.bucketArn`).
	//
	// Normally, this token will resolve to `arnAttr`, but if the resource is
	// referenced across environments, `arnComponents` will be used to synthesize
	// a concrete ARN with the resource's physical name. Make sure to reference
	// `this.physicalName` in `arnComponents`.
	// Experimental.
	GetResourceArnAttribute(arnAttr *string, arnComponents *awscdk.ArnComponents) *string
	// Returns an environment-sensitive token that should be used for the resource's "name" attribute (e.g. `bucket.bucketName`).
	//
	// Normally, this token will resolve to `nameAttr`, but if the resource is
	// referenced across environments, it will be resolved to `this.physicalName`,
	// which will be a concrete name.
	// Experimental.
	GetResourceNameAttribute(nameAttr *string) *string
	// Grant the given principal identity permissions to perform actions on this guardrail.
	//
	// [disable-awslint:no-grants].
	// Experimental.
	Grant(grantee awsiam.IGrantable, actions ...*string) awsiam.Grant
	// Grant the given identity permissions to apply the guardrail.
	//
	// [disable-awslint:no-grants].
	// Experimental.
	GrantApply(grantee awsiam.IGrantable) awsiam.Grant
	// Return the given named metric for this guardrail.
	//
	// By default, the metric will be calculated as a sum over a period of 5 minutes.
	// You can customize this by using the `statistic` and `period` properties.
	// Experimental.
	Metric(metricName *string, props *awscloudwatch.MetricOptions) awscloudwatch.Metric
	// Return the invocation client errors metric for this guardrail.
	// Experimental.
	MetricInvocationClientErrors(props *awscloudwatch.MetricOptions) awscloudwatch.Metric
	// Return the invocation latency metric for this guardrail.
	// Experimental.
	MetricInvocationLatency(props *awscloudwatch.MetricOptions) awscloudwatch.Metric
	// Return the invocations metric for this guardrail.
	// Experimental.
	MetricInvocations(props *awscloudwatch.MetricOptions) awscloudwatch.Metric
	// Return the invocation server errors metric for this guardrail.
	// Experimental.
	MetricInvocationServerErrors(props *awscloudwatch.MetricOptions) awscloudwatch.Metric
	// Return the invocations intervened metric for this guardrail.
	// Experimental.
	MetricInvocationsIntervened(props *awscloudwatch.MetricOptions) awscloudwatch.Metric
	// Return the invocation throttles metric for this guardrail.
	// Experimental.
	MetricInvocationThrottles(props *awscloudwatch.MetricOptions) awscloudwatch.Metric
	// Return the text unit count metric for this guardrail.
	// Experimental.
	MetricTextUnitCount(props *awscloudwatch.MetricOptions) awscloudwatch.Metric
	// Returns a string representation of this construct.
	// Experimental.
	ToString() *string
	// Experimental.
	UpdateVersion(version *string)
	// Applies one or more mixins to this construct.
	//
	// Mixins are applied in order. The list of constructs is captured at the
	// start of the call, so constructs added by a mixin will not be visited.
	// Use multiple `with()` calls if subsequent mixins should apply to added
	// constructs.
	// Experimental.
	With(mixins ...constructs.IMixin) constructs.IConstruct
}

Abstract base class for a Guardrail.

Contains methods and attributes valid for Guardrails either created with CDK or imported. Experimental.

type GuardrailCrossRegionConfigProperty 

type GuardrailCrossRegionConfigProperty struct {
	// The arn of thesystem-defined guardrail profile that you're using with your guardrail.
	//
	// Guardrail profiles define the destination AWS Regions where guardrail inference requests can be automatically routed.
	// Using guardrail profiles helps maintain guardrail performance and reliability when demand increases.
	// Default: - No cross-region configuration.
	//
	// Experimental.
	GuardrailProfileArn *string `field:"required" json:"guardrailProfileArn" yaml:"guardrailProfileArn"`
}

GuardrailCrossRegionConfigProperty.

Example: 

guardrail := bedrock.NewGuardrail(this, jsii.String("bedrockGuardrails"), &GuardrailProps{
	GuardrailName: jsii.String("my-BedrockGuardrails"),
	Description: jsii.String("Guardrail with cross-region configuration for enhanced language support"),
	CrossRegionConfig: &GuardrailCrossRegionConfigProperty{
		GuardrailProfileArn: jsii.String("arn:aws:bedrock:us-east-1:123456789012:guardrail-profile/my-profile"),
	},
	// Use STANDARD tier for enhanced capabilities
	ContentFiltersTierConfig: bedrock.TierConfig_STANDARD,
	TopicsTierConfig: bedrock.TierConfig_STANDARD,
})

Experimental.

type GuardrailProps 

type GuardrailProps struct {
	// The name of the guardrail.
	//
	// This will be used as the physical name of the guardrail.
	// Experimental.
	GuardrailName *string `field:"required" json:"guardrailName" yaml:"guardrailName"`
	// The message to return when the guardrail blocks a prompt.
	//
	// Must be between 1 and 500 characters.
	// Default: "Sorry, your query violates our usage policy."
	//
	// Experimental.
	BlockedInputMessaging *string `field:"optional" json:"blockedInputMessaging" yaml:"blockedInputMessaging"`
	// The message to return when the guardrail blocks a model response.
	//
	// Must be between 1 and 500 characters.
	// Default: "Sorry, I am unable to answer your question because of our usage policy."
	//
	// Experimental.
	BlockedOutputsMessaging *string `field:"optional" json:"blockedOutputsMessaging" yaml:"blockedOutputsMessaging"`
	// The content filters to apply to the guardrail.
	// Default: [].
	//
	// Experimental.
	ContentFilters *[]*ContentFilter `field:"optional" json:"contentFilters" yaml:"contentFilters"`
	// The tier configuration to apply to the guardrail.
	// Default: filters.TierConfig.CLASSIC
	//
	// Experimental.
	ContentFiltersTierConfig TierConfig `field:"optional" json:"contentFiltersTierConfig" yaml:"contentFiltersTierConfig"`
	// The contextual grounding filters to apply to the guardrail.
	// Default: [].
	//
	// Experimental.
	ContextualGroundingFilters *[]*ContextualGroundingFilter `field:"optional" json:"contextualGroundingFilters" yaml:"contextualGroundingFilters"`
	// The cross-region configuration for the guardrail.
	//
	// This is optional and when provided, it should be of type GuardrailCrossRegionConfigProperty.
	// Default: - No cross-region configuration.
	//
	// Experimental.
	CrossRegionConfig *GuardrailCrossRegionConfigProperty `field:"optional" json:"crossRegionConfig" yaml:"crossRegionConfig"`
	// A list of policies related to topics that the guardrail should deny.
	// Default: [].
	//
	// Experimental.
	DeniedTopics *[]Topic `field:"optional" json:"deniedTopics" yaml:"deniedTopics"`
	// The description of the guardrail.
	// Default: - No description.
	//
	// Experimental.
	Description *string `field:"optional" json:"description" yaml:"description"`
	// A custom KMS key to use for encrypting data.
	// Default: - Data is encrypted by default with a key that AWS owns and manages for you.
	//
	// Experimental.
	KmsKey awskms.IKey `field:"optional" json:"kmsKey" yaml:"kmsKey"`
	// The managed word filters to apply to the guardrail.
	// Default: [].
	//
	// Experimental.
	ManagedWordListFilters *[]*ManagedWordFilter `field:"optional" json:"managedWordListFilters" yaml:"managedWordListFilters"`
	// The PII filters to apply to the guardrail.
	// Default: [].
	//
	// Experimental.
	PiiFilters *[]*PIIFilter `field:"optional" json:"piiFilters" yaml:"piiFilters"`
	// The regular expression (regex) filters to apply to the guardrail.
	// Default: [].
	//
	// Experimental.
	RegexFilters *[]*RegexFilter `field:"optional" json:"regexFilters" yaml:"regexFilters"`
	// The tier configuration to apply to the guardrail.
	// Default: filters.TierConfig.CLASSIC
	//
	// Experimental.
	TopicsTierConfig TierConfig `field:"optional" json:"topicsTierConfig" yaml:"topicsTierConfig"`
	// The word filters to apply to the guardrail.
	// Default: [].
	//
	// Experimental.
	WordFilters *[]*WordFilter `field:"optional" json:"wordFilters" yaml:"wordFilters"`
}

Properties for creating a Guardrail.

Example: 

guardrail := bedrock.NewGuardrail(this, jsii.String("bedrockGuardrails"), &GuardrailProps{
	GuardrailName: jsii.String("my-BedrockGuardrails"),
})
// Add regex filter with input/output actions
guardrail.AddRegexFilter(&RegexFilter{
	Name: jsii.String("TestRegexFilter"),
	Pattern: jsii.String("test-pattern"),
	Action: bedrock.GuardrailAction_ANONYMIZE,
	// below props are optional
	Description: jsii.String("This is a test regex filter"),
	InputAction: bedrock.GuardrailAction_BLOCK,
	InputEnabled: jsii.Boolean(true),
	OutputAction: bedrock.GuardrailAction_ANONYMIZE,
	OutputEnabled: jsii.Boolean(true),
})

Experimental.

type IAgent 

type IAgent interface {
	awsiam.IGrantable
	awscdk.IResource
	// Grant invoke permissions on this agent to an IAM principal.
	//
	// Note: This grant will only work when the grantee is in the same AWS account
	// where the agent is defined. Cross-account invocation is not supported.
	// Experimental.
	GrantInvoke(grantee awsiam.IGrantable) awsiam.Grant
	// Return the CloudWatch metric for agent count.
	// Experimental.
	MetricCount(props *awscloudwatch.MetricOptions) awscloudwatch.Metric
	// Defines a CloudWatch event rule triggered by agent events.
	// Experimental.
	OnEvent(id *string, options *awsevents.OnEventOptions) awsevents.Rule
	// The ARN of the agent.
	// Experimental.
	AgentArn() *string
	// The ID of the Agent.
	// Experimental.
	AgentId() *string
	// Optional KMS encryption key associated with this agent.
	// Experimental.
	KmsKey() awskms.IKey
	// When this agent was last updated.
	// Experimental.
	LastUpdated() *string
	// The IAM role associated to the agent.
	// Experimental.
	Role() awsiam.IRole
}

Represents an Agent, either created with CDK or imported. Experimental.

func Agent_FromAgentAttributes 

func Agent_FromAgentAttributes(scope constructs.Construct, id *string, attrs *AgentAttributes) IAgent

Creates an Agent reference from an existing agent's attributes.

Returns: An IAgent reference to the existing agent. Default: - For attrs.agentVersion: 'DRAFT' if no explicit version is provided

Experimental.

type IAgentAlias 

type IAgentAlias interface {
	awscdk.IResource
	// Grant the given principal identity permissions to perform actions on this agent alias.
	// Experimental.
	Grant(grantee awsiam.IGrantable, actions ...*string) awsiam.Grant
	// Grant the given identity permissions to get the agent alias.
	// Experimental.
	GrantGet(grantee awsiam.IGrantable) awsiam.Grant
	// Grant the given identity permissions to invoke the agent alias.
	// Experimental.
	GrantInvoke(grantee awsiam.IGrantable) awsiam.Grant
	// Define an EventBridge rule that triggers when something happens to this agent alias.
	//
	// Requires that there exists at least one CloudTrail Trail in your account
	// that captures the event. This method will not create the Trail.
	// Experimental.
	OnCloudTrailEvent(id *string, options *awsevents.OnEventOptions) awsevents.Rule
	// The underlying agent for this alias.
	// Experimental.
	Agent() IAgent
	// The ARN of the agent alias.
	// Experimental.
	AliasArn() *string
	// The unique identifier of the agent alias.
	// Experimental.
	AliasId() *string
}

Represents an Agent Alias, either created with CDK or imported. Experimental.

func AgentAlias_FromAttributes 

func AgentAlias_FromAttributes(scope constructs.Construct, id *string, attrs *AgentAliasAttributes) IAgentAlias

Brings an Agent Alias from an existing one created outside of CDK. Experimental.

type IBedrockInvokable 

type IBedrockInvokable interface {
	// Gives the appropriate policies to invoke and use the invokable abstraction.
	// Experimental.
	GrantInvoke(grantee awsiam.IGrantable) awsiam.Grant
	// The ARN of the Bedrock invokable abstraction.
	// Experimental.
	InvokableArn() *string
}

Represents an Amazon Bedrock abstraction on which you can run the `Invoke` API.

This can be a Foundational Model, a Custom Model, or an Inference Profile. Experimental.

type IGuardrail 

type IGuardrail interface {
	awscdk.IResource
	// Grant the given principal identity permissions to perform actions on this guardrail.
	// Experimental.
	Grant(grantee awsiam.IGrantable, actions ...*string) awsiam.Grant
	// Grant the given identity permissions to apply the guardrail.
	// Experimental.
	GrantApply(grantee awsiam.IGrantable) awsiam.Grant
	// Return the given named metric for this guardrail.
	// Experimental.
	Metric(metricName *string, props *awscloudwatch.MetricOptions) awscloudwatch.Metric
	// Return the invocation client errors metric for this guardrail.
	// Experimental.
	MetricInvocationClientErrors(props *awscloudwatch.MetricOptions) awscloudwatch.Metric
	// Return the invocation latency metric for this guardrail.
	// Experimental.
	MetricInvocationLatency(props *awscloudwatch.MetricOptions) awscloudwatch.Metric
	// Return the invocations metric for this guardrail.
	// Experimental.
	MetricInvocations(props *awscloudwatch.MetricOptions) awscloudwatch.Metric
	// Return the invocation server errors metric for this guardrail.
	// Experimental.
	MetricInvocationServerErrors(props *awscloudwatch.MetricOptions) awscloudwatch.Metric
	// Return the invocations intervened metric for this guardrail.
	// Experimental.
	MetricInvocationsIntervened(props *awscloudwatch.MetricOptions) awscloudwatch.Metric
	// Return the invocation throttles metric for this guardrail.
	// Experimental.
	MetricInvocationThrottles(props *awscloudwatch.MetricOptions) awscloudwatch.Metric
	// Return the text unit count metric for this guardrail.
	// Experimental.
	MetricTextUnitCount(props *awscloudwatch.MetricOptions) awscloudwatch.Metric
	// The ARN of the guardrail.
	// Experimental.
	GuardrailArn() *string
	// The ID of the guardrail.
	// Experimental.
	GuardrailId() *string
	// The version of the guardrail.
	//
	// If no explicit version is created,
	// this will default to "DRAFT".
	// Experimental.
	GuardrailVersion() *string
	// Optional KMS encryption key associated with this guardrail.
	// Experimental.
	KmsKey() awskms.IKey
	// When this guardrail was last updated.
	// Experimental.
	LastUpdated() *string
}

Represents a Guardrail, either created with CDK or imported. Experimental.

func Guardrail_FromCfnGuardrail 

func Guardrail_FromCfnGuardrail(cfnGuardrail awsbedrock.CfnGuardrail) IGuardrail

Import a low-level L1 Cfn Guardrail. Experimental.

func Guardrail_FromGuardrailAttributes 

func Guardrail_FromGuardrailAttributes(scope constructs.Construct, id *string, attrs *GuardrailAttributes) IGuardrail

Import a guardrail given its attributes. Experimental.

type IInferenceProfile 

type IInferenceProfile interface {
	// Grants appropriate permissions to use the inference profile.
	//
	// Each profile type requires different permissions based on its usage pattern.
	//
	// Returns: An IAM Grant object representing the granted permissions.
	// Experimental.
	GrantProfileUsage(grantee awsiam.IGrantable) awsiam.Grant
	// The ARN of the inference profile.
	// Experimental.
	InferenceProfileArn() *string
	// The unique identifier of the inference profile.
	// Experimental.
	InferenceProfileId() *string
	// The type of inference profile.
	// Experimental.
	Type() InferenceProfileType
}

Represents an Inference Profile, either created with CDK or imported. Experimental.

func ApplicationInferenceProfile_FromApplicationInferenceProfileAttributes 

func ApplicationInferenceProfile_FromApplicationInferenceProfileAttributes(scope constructs.Construct, id *string, attrs *ApplicationInferenceProfileAttributes) IInferenceProfile

Import an Application Inference Profile given its attributes.

[disable-awslint:no-grants].

Returns: An IInferenceProfile reference to the existing application inference profile. Experimental.

func ApplicationInferenceProfile_FromCfnApplicationInferenceProfile 

func ApplicationInferenceProfile_FromCfnApplicationInferenceProfile(cfnApplicationInferenceProfile awsbedrock.CfnApplicationInferenceProfile) IInferenceProfile

Import a low-level L1 Cfn Application Inference Profile.

[disable-awslint:no-grants].

Returns: An IInferenceProfile reference to the imported application inference profile. Experimental.

type IPrompt 

type IPrompt interface {
	awscdk.IResource
	// Grant the given identity permissions to get the prompt.
	// Experimental.
	GrantGet(grantee awsiam.IGrantable) awsiam.Grant
	// Optional KMS encryption key associated with this prompt.
	// Experimental.
	KmsKey() awskms.IKey
	// The ARN of the prompt.
	//
	// Example:
	//   "arn:aws:bedrock:us-east-1:123456789012:prompt/PROMPT12345"
	//
	// Experimental.
	PromptArn() *string
	// The ID of the prompt.
	//
	// Example:
	//   "PROMPT12345"
	//
	// Experimental.
	PromptId() *string
	// The version of the prompt.
	// Default: "DRAFT".
	//
	// Experimental.
	PromptVersion() *string
}

Represents a Prompt, either created with CDK or imported. Experimental.

func Prompt_FromPromptAttributes 

func Prompt_FromPromptAttributes(scope constructs.Construct, id *string, attrs *PromptAttributes) IPrompt

Creates a Prompt reference from an existing prompt's attributes.

Returns: An IPrompt reference to the existing prompt. Default: - For attrs.promptVersion: 'DRAFT' if no explicit version is provided

Experimental.

type IPromptRouter 

type IPromptRouter interface {
	// The ARN of the prompt router.
	// Experimental.
	PromptRouterArn() *string
	// The ID of the prompt router.
	// Experimental.
	PromptRouterId() *string
	// The foundation models / profiles this router will route to.
	// Experimental.
	RoutingEndpoints() *[]IBedrockInvokable
}

Represents a Prompt Router, which provides intelligent routing between different models. Experimental.

type IPromptVariant 

type IPromptVariant interface {
	// The generative AI resource configuration.
	// Experimental.
	GenAiResource() PromptGenAiResource
	// The inference configuration.
	// Experimental.
	InferenceConfiguration() PromptInferenceConfiguration
	// The unique identifier of the model with which to run inference on the prompt.
	// Experimental.
	ModelId() *string
	// The name of the prompt variant.
	// Experimental.
	Name() *string
	// The template configuration.
	// Experimental.
	TemplateConfiguration() PromptTemplateConfiguration
	// The type of prompt template.
	// Default: - Text.
	//
	// Experimental.
	TemplateType() PromptTemplateType
}

Interface representing a prompt variant configuration.

Variants are specific sets of inputs that guide FMs on Amazon Bedrock to generate an appropriate response or output for a given task or instruction. Experimental.

func PromptVariant_Agent 

func PromptVariant_Agent(props *AgentPromptVariantProps) IPromptVariant

Creates an agent prompt template variant.

Returns: A PromptVariant configured for agent interactions. Experimental.

func PromptVariant_Chat 

func PromptVariant_Chat(props *ChatPromptVariantProps) IPromptVariant

Creates a chat template variant.

Use this template type when the model supports the Converse API or the Anthropic Claude Messages API. This allows you to include a System prompt and previous User messages and Assistant messages for context.

Returns: A PromptVariant configured for chat interactions. Experimental.

func PromptVariant_Text 

func PromptVariant_Text(props *TextPromptVariantProps) IPromptVariant

Creates a text template variant.

Returns: A PromptVariant configured for text processing. Experimental.

type InferenceConfiguration 

type InferenceConfiguration struct {
	// The maximum number of tokens to generate in the response.
	//
	// Integer
	//
	// min 0
	// max 4096.
	// Experimental.
	MaximumLength *float64 `field:"required" json:"maximumLength" yaml:"maximumLength"`
	// A list of stop sequences.
	//
	// A stop sequence is a sequence of characters that
	// causes the model to stop generating the response.
	//
	// length 0-4.
	// Experimental.
	StopSequences *[]*string `field:"required" json:"stopSequences" yaml:"stopSequences"`
	// The likelihood of the model selecting higher-probability options while generating a response.
	//
	// A lower value makes the model more likely to choose
	// higher-probability options, while a higher value makes the model more
	// likely to choose lower-probability options.
	//
	// Floating point
	//
	// min 0
	// max 1.
	// Experimental.
	Temperature *float64 `field:"required" json:"temperature" yaml:"temperature"`
	// While generating a response, the model determines the probability of the following token at each point of generation.
	//
	// The value that you set for
	// topK is the number of most-likely candidates from which the model chooses
	// the next token in the sequence. For example, if you set topK to 50, the
	// model selects the next token from among the top 50 most likely choices.
	//
	// Integer
	//
	// min 0
	// max 500.
	// Experimental.
	TopK *float64 `field:"required" json:"topK" yaml:"topK"`
	// While generating a response, the model determines the probability of the following token at each point of generation.
	//
	// The value that you set for
	// Top P determines the number of most-likely candidates from which the model
	// chooses the next token in the sequence. For example, if you set topP to
	// 80, the model only selects the next token from the top 80% of the
	// probability distribution of next tokens.
	//
	// Floating point
	//
	// min 0
	// max 1.
	// Experimental.
	TopP *float64 `field:"required" json:"topP" yaml:"topP"`
}

LLM inference configuration.

Example: 

agent := bedrock.NewAgent(this, jsii.String("Agent"), &AgentProps{
	FoundationModel: bedrock.BedrockFoundationModel_AMAZON_NOVA_LITE_V1(),
	Instruction: jsii.String("You are a helpful assistant."),
	PromptOverrideConfiguration: bedrock.PromptOverrideConfiguration_FromSteps([]PromptStepConfigBase{
		&PromptStepConfigBase{
			StepType: bedrock.AgentStepType_PRE_PROCESSING,
			StepEnabled: jsii.Boolean(true),
			CustomPromptTemplate: jsii.String("Your custom prompt template here"),
			InferenceConfig: &InferenceConfiguration{
				Temperature: jsii.Number(0),
				TopP: jsii.Number(1),
				TopK: jsii.Number(250),
				MaximumLength: jsii.Number(1),
				StopSequences: []*string{
					jsii.String("\n\nHuman:"),
				},
			},
		},
	}),
})

Experimental.

type InferenceProfileBase 

type InferenceProfileBase interface {
	awscdk.Resource
	IInferenceProfile
	// The environment this resource belongs to.
	//
	// For resources that are created and managed in a Stack (those created by
	// creating new class instances like `new Role()`, `new Bucket()`, etc.), this
	// is always the same as the environment of the stack they belong to.
	//
	// For referenced resources (those obtained from referencing methods like
	// `Role.fromRoleArn()`, `Bucket.fromBucketName()`, etc.), they might be
	// different than the stack they were imported into.
	// Experimental.
	Env() *interfaces.ResourceEnvironment
	// The ARN of the inference profile.
	// Experimental.
	InferenceProfileArn() *string
	// The unique identifier of the inference profile.
	// Experimental.
	InferenceProfileId() *string
	// The tree node.
	// Experimental.
	Node() constructs.Node
	// Returns a string-encoded token that resolves to the physical name that should be passed to the CloudFormation resource.
	//
	// This value will resolve to one of the following:
	// - a concrete value (e.g. `"my-awesome-bucket"`)
	// - `undefined`, when a name should be generated by CloudFormation
	// - a concrete name generated automatically during synthesis, in
	//   cross-environment scenarios.
	// Experimental.
	PhysicalName() *string
	// The stack in which this resource is defined.
	// Experimental.
	Stack() awscdk.Stack
	// The type of inference profile (SYSTEM_DEFINED or APPLICATION).
	// Experimental.
	Type() InferenceProfileType
	// Apply the given removal policy to this resource.
	//
	// The Removal Policy controls what happens to this resource when it stops
	// being managed by CloudFormation, either because you've removed it from the
	// CDK application or because you've made a change that requires the resource
	// to be replaced.
	//
	// The resource can be deleted (`RemovalPolicy.DESTROY`), or left in your AWS
	// account for data recovery and cleanup later (`RemovalPolicy.RETAIN`).
	// Experimental.
	ApplyRemovalPolicy(policy awscdk.RemovalPolicy)
	// Experimental.
	GeneratePhysicalName() *string
	// Returns an environment-sensitive token that should be used for the resource's "ARN" attribute (e.g. `bucket.bucketArn`).
	//
	// Normally, this token will resolve to `arnAttr`, but if the resource is
	// referenced across environments, `arnComponents` will be used to synthesize
	// a concrete ARN with the resource's physical name. Make sure to reference
	// `this.physicalName` in `arnComponents`.
	// Experimental.
	GetResourceArnAttribute(arnAttr *string, arnComponents *awscdk.ArnComponents) *string
	// Returns an environment-sensitive token that should be used for the resource's "name" attribute (e.g. `bucket.bucketName`).
	//
	// Normally, this token will resolve to `nameAttr`, but if the resource is
	// referenced across environments, it will be resolved to `this.physicalName`,
	// which will be a concrete name.
	// Experimental.
	GetResourceNameAttribute(nameAttr *string) *string
	// Grants appropriate permissions to use the inference profile.
	//
	// Each profile type requires different permissions based on its usage pattern:
	// - Application profiles need bedrock:InvokeModel for direct invocation
	// - Cross-region profiles need bedrock:InvokeModel* for routing capabilities
	//
	// Note: This does not grant permissions to use the underlying model in the profile.
	// For comprehensive model invocation permissions, use the grantInvoke method instead.
	//
	// [disable-awslint:no-grants].
	//
	// Returns: An IAM Grant object representing the granted permissions.
	// Experimental.
	GrantProfileUsage(grantee awsiam.IGrantable) awsiam.Grant
	// Returns a string representation of this construct.
	// Experimental.
	ToString() *string
	// Applies one or more mixins to this construct.
	//
	// Mixins are applied in order. The list of constructs is captured at the
	// start of the call, so constructs added by a mixin will not be visited.
	// Use multiple `with()` calls if subsequent mixins should apply to added
	// constructs.
	// Experimental.
	With(mixins ...constructs.IMixin) constructs.IConstruct
}

Abstract base class for an Inference Profile.

Contains methods and attributes valid for Inference Profiles either created with CDK or imported. Experimental.

type InferenceProfileType 

type InferenceProfileType string

These are the values used by the API when using aws bedrock get-inference-profile --inference-profile-identifier XXXXXXX. Experimental. 

const (
	// An inference profile that is created by AWS.
	//
	// These are profiles such as cross-region
	// which help you distribute traffic across a geographic region.
	// Experimental.
	InferenceProfileType_SYSTEM_DEFINED InferenceProfileType = "SYSTEM_DEFINED"
	// An inference profile that is user-created.
	//
	// These are profiles that help
	// you track costs or metrics.
	// Experimental.
	InferenceProfileType_APPLICATION InferenceProfileType = "APPLICATION"
)

type InformationTechnologyPIIType 

type InformationTechnologyPIIType interface {
	PIIType
	// The string value of the PII type.
	// Experimental.
	Value() *string
	// Returns the string representation of the PII type.
	//
	// Returns: The string value of the PII type.
	// Experimental.
	ToString() *string
}

Types of PII in the domain of IT (Information Technology).

Example: 

guardrail := bedrock.NewGuardrail(this, jsii.String("bedrockGuardrails"), &GuardrailProps{
	GuardrailName: jsii.String("my-BedrockGuardrails"),
})

// Add PII filter for addresses with input/output actions
guardrail.AddPIIFilter(&PIIFilter{
	Type: bedrock.GeneralPIIType_ADDRESS(),
	Action: bedrock.GuardrailAction_BLOCK,
	// below props are optional
	InputAction: bedrock.GuardrailAction_BLOCK,
	InputEnabled: jsii.Boolean(true),
	OutputAction: bedrock.GuardrailAction_ANONYMIZE,
	OutputEnabled: jsii.Boolean(true),
})

// Add PII filter for credit card numbers with input/output actions
guardrail.AddPIIFilter(&PIIFilter{
	Type: bedrock.FinancePIIType_CREDIT_DEBIT_CARD_NUMBER(),
	Action: bedrock.GuardrailAction_BLOCK,
	// below props are optional
	InputAction: bedrock.GuardrailAction_BLOCK,
	InputEnabled: jsii.Boolean(true),
	OutputAction: bedrock.GuardrailAction_ANONYMIZE,
	OutputEnabled: jsii.Boolean(true),
})

// Add PII filter for email addresses
guardrail.AddPIIFilter(&PIIFilter{
	Type: bedrock.GeneralPIIType_EMAIL(),
	Action: bedrock.GuardrailAction_ANONYMIZE,
})

// Add PII filter for US Social Security Numbers
guardrail.AddPIIFilter(&PIIFilter{
	Type: bedrock.USASpecificPIIType_US_SOCIAL_SECURITY_NUMBER(),
	Action: bedrock.GuardrailAction_BLOCK,
})

// Add PII filter for IP addresses
guardrail.AddPIIFilter(&PIIFilter{
	Type: bedrock.InformationTechnologyPIIType_IP_ADDRESS(),
	Action: bedrock.GuardrailAction_ANONYMIZE,
})

Experimental.

func InformationTechnologyPIIType_AWS_ACCESS_KEY 

func InformationTechnologyPIIType_AWS_ACCESS_KEY() InformationTechnologyPIIType

func InformationTechnologyPIIType_AWS_SECRET_KEY 

func InformationTechnologyPIIType_AWS_SECRET_KEY() InformationTechnologyPIIType

func InformationTechnologyPIIType_IP_ADDRESS 

func InformationTechnologyPIIType_IP_ADDRESS() InformationTechnologyPIIType

func InformationTechnologyPIIType_MAC_ADDRESS 

func InformationTechnologyPIIType_MAC_ADDRESS() InformationTechnologyPIIType

func InformationTechnologyPIIType_URL 

func InformationTechnologyPIIType_URL() InformationTechnologyPIIType

type InlineApiSchema 

type InlineApiSchema interface {
	ApiSchema
	// The inline OpenAPI schema definition as a string, if using an inline schema.
	//
	// Can be in JSON or YAML format.
	// Experimental.
	InlineSchema() *string
	// The S3 location of the API schema file, if using an S3-based schema.
	//
	// Contains the bucket name and object key information.
	// Experimental.
	S3File() *awss3.Location
}

Class to define an API Schema from an inline string.

The schema can be provided directly as a string in either JSON or YAML format.

Example: 

inlineSchema := bedrock.ApiSchema_FromInline(jsii.String(`
openapi: 3.0.3
info:
  title: Library API
  version: 1.0.0
paths:
  /search:
    get:
      summary: Search for books
      operationId: searchBooks
      parameters:
        - name: query
          in: query
          required: true
          schema:
            type: string
`))

actionGroupFunction := lambda.NewFunction(this, jsii.String("ActionGroupFunction"), &FunctionProps{
	Runtime: lambda.Runtime_PYTHON_3_12(),
	Handler: jsii.String("index.handler"),
	Code: lambda.Code_FromAsset(path.join(__dirname, jsii.String("../lambda/action-group"))),
})

actionGroup := bedrock.NewAgentActionGroup(&AgentActionGroupProps{
	Name: jsii.String("query-library"),
	Description: jsii.String("Use these functions to get information about the books in the library."),
	Executor: bedrock.ActionGroupExecutor_FromLambda(actionGroupFunction),
	Enabled: jsii.Boolean(true),
	ApiSchema: inlineSchema,
})

agent := bedrock.NewAgent(this, jsii.String("Agent"), &AgentProps{
	FoundationModel: bedrock.BedrockFoundationModel_ANTHROPIC_CLAUDE_HAIKU_V1_0(),
	Instruction: jsii.String("You are a helpful and friendly agent that answers questions about literature."),
})

agent.AddActionGroup(actionGroup)

Experimental.

func ApiSchema_FromInline 

func ApiSchema_FromInline(schema *string) InlineApiSchema

Creates an API Schema from an inline string. Experimental.

func AssetApiSchema_FromInline 

func AssetApiSchema_FromInline(schema *string) InlineApiSchema

Creates an API Schema from an inline string. Experimental.

func InlineApiSchema_FromInline 

func InlineApiSchema_FromInline(schema *string) InlineApiSchema

Creates an API Schema from an inline string. Experimental.

func NewInlineApiSchema 

func NewInlineApiSchema(schema *string) InlineApiSchema

Experimental.

func S3ApiSchema_FromInline 

func S3ApiSchema_FromInline(schema *string) InlineApiSchema

Creates an API Schema from an inline string. Experimental.

type ManagedWordFilter 

type ManagedWordFilter struct {
	// The action to take when a managed word is detected in the input.
	// Default: GuardrailAction.BLOCK
	//
	// Experimental.
	InputAction GuardrailAction `field:"optional" json:"inputAction" yaml:"inputAction"`
	// Whether the managed word filter is enabled for input.
	// Default: true.
	//
	// Experimental.
	InputEnabled *bool `field:"optional" json:"inputEnabled" yaml:"inputEnabled"`
	// The action to take when a managed word is detected in the output.
	// Default: GuardrailAction.BLOCK
	//
	// Experimental.
	OutputAction GuardrailAction `field:"optional" json:"outputAction" yaml:"outputAction"`
	// Whether the managed word filter is enabled for output.
	// Default: true.
	//
	// Experimental.
	OutputEnabled *bool `field:"optional" json:"outputEnabled" yaml:"outputEnabled"`
	// The type of managed word filter.
	// Default: ManagedWordFilterType.PROFANITY
	//
	// Experimental.
	Type ManagedWordFilterType `field:"optional" json:"type" yaml:"type"`
}

Interface for managed word list filters.

Example: 

guardrail := bedrock.NewGuardrail(this, jsii.String("bedrockGuardrails"), &GuardrailProps{
	GuardrailName: jsii.String("my-BedrockGuardrails"),
})

// Add managed word list with input/output actions
guardrail.AddManagedWordListFilter(&ManagedWordFilter{
	Type: bedrock.ManagedWordFilterType_PROFANITY,
	InputAction: bedrock.GuardrailAction_BLOCK,
	InputEnabled: jsii.Boolean(true),
	OutputAction: bedrock.GuardrailAction_NONE,
	OutputEnabled: jsii.Boolean(true),
})

// Add individual words
guardrail.AddWordFilter(&WordFilter{
	Text: jsii.String("drugs"),
})
guardrail.AddWordFilter(&WordFilter{
	Text: jsii.String("competitor"),
})

// Add words from a file
guardrail.AddWordFilterFromFile(jsii.String("./scripts/wordsPolicy.csv"))

Experimental.

type ManagedWordFilterType 

type ManagedWordFilterType string

Managed word list filter types supported by Amazon Bedrock.

Example: 

guardrail := bedrock.NewGuardrail(this, jsii.String("bedrockGuardrails"), &GuardrailProps{
	GuardrailName: jsii.String("my-BedrockGuardrails"),
})

// Add managed word list with input/output actions
guardrail.AddManagedWordListFilter(&ManagedWordFilter{
	Type: bedrock.ManagedWordFilterType_PROFANITY,
	InputAction: bedrock.GuardrailAction_BLOCK,
	InputEnabled: jsii.Boolean(true),
	OutputAction: bedrock.GuardrailAction_NONE,
	OutputEnabled: jsii.Boolean(true),
})

// Add individual words
guardrail.AddWordFilter(&WordFilter{
	Text: jsii.String("drugs"),
})
guardrail.AddWordFilter(&WordFilter{
	Text: jsii.String("competitor"),
})

// Add words from a file
guardrail.AddWordFilterFromFile(jsii.String("./scripts/wordsPolicy.csv"))

Experimental. 

const (
	// Filter for profanity and explicit language.
	// Experimental.
	ManagedWordFilterType_PROFANITY ManagedWordFilterType = "PROFANITY"
)

type Memory 

type Memory interface {
}

Memory class for managing Bedrock Agent memory configurations.

Enables conversational context retention across multiple sessions through session identifiers. Memory context is stored with unique memory IDs per user, allowing access to conversation history and summaries. Supports viewing stored sessions and clearing memory.

Example: 

agent := bedrock.NewAgent(this, jsii.String("MyAgent"), &AgentProps{
	AgentName: jsii.String("MyAgent"),
	Instruction: jsii.String("Your instruction here"),
	FoundationModel: bedrock.BedrockFoundationModel_AMAZON_NOVA_LITE_V1(),
	Memory: awsbedrockalpha.Memory_SessionSummary(&SessionSummaryMemoryProps{
		MaxRecentSessions: jsii.Number(10),
		 // Keep the last 10 session summaries
		MemoryDuration: awscdk.Duration_Days(jsii.Number(20)),
	}),
})

See: https://docs.aws.amazon.com/bedrock/latest/userguide/agents-memory.html

Experimental.

func Memory_SESSION_SUMMARY 

func Memory_SESSION_SUMMARY() Memory

func Memory_SessionSummary 

func Memory_SessionSummary(props *SessionSummaryMemoryProps) Memory

Creates a session summary memory with custom configuration.

Returns: Memory instance. Experimental.

func NewMemory 

func NewMemory(props *SessionSummaryMemoryProps) Memory

Experimental.

type ModalityType 

type ModalityType string

The type of modality that can be used in content filters.

Example: 

guardrail := bedrock.NewGuardrail(this, jsii.String("bedrockGuardrails"), &GuardrailProps{
	GuardrailName: jsii.String("my-BedrockGuardrails"),
	// Configure tier for content filters (optional)
	ContentFiltersTierConfig: bedrock.TierConfig_STANDARD,
})

guardrail.AddContentFilter(&ContentFilter{
	Type: bedrock.ContentFilterType_SEXUAL,
	InputStrength: bedrock.ContentFilterStrength_HIGH,
	OutputStrength: bedrock.ContentFilterStrength_MEDIUM,
	// props below are optional
	InputAction: bedrock.GuardrailAction_BLOCK,
	InputEnabled: jsii.Boolean(true),
	OutputAction: bedrock.GuardrailAction_NONE,
	OutputEnabled: jsii.Boolean(true),
	InputModalities: []ModalityType{
		bedrock.ModalityType_TEXT,
		bedrock.ModalityType_IMAGE,
	},
	OutputModalities: []ModalityType{
		bedrock.ModalityType_TEXT,
	},
})

Experimental. 

const (
	// Text modality for content filters.
	// Experimental.
	ModalityType_TEXT ModalityType = "TEXT"
	// Image modality for content filters.
	// Experimental.
	ModalityType_IMAGE ModalityType = "IMAGE"
)

type OrchestrationType 

type OrchestrationType string

Enum for orchestration types available for agents. Experimental. 

const (
	// Default orchestration by the agent.
	// Experimental.
	OrchestrationType_DEFAULT OrchestrationType = "DEFAULT"
	// Custom orchestration using Lambda.
	// Experimental.
	OrchestrationType_CUSTOM_ORCHESTRATION OrchestrationType = "CUSTOM_ORCHESTRATION"
)

type PIIFilter 

type PIIFilter struct {
	// The action to take when PII is detected.
	// Experimental.
	Action GuardrailAction `field:"required" json:"action" yaml:"action"`
	// The type of PII to filter.
	// Experimental.
	Type PIIType `field:"required" json:"type" yaml:"type"`
	// The action to take when PII is detected in the input.
	// Default: GuardrailAction.BLOCK
	//
	// Experimental.
	InputAction GuardrailAction `field:"optional" json:"inputAction" yaml:"inputAction"`
	// Whether the PII filter is enabled for input.
	// Default: true.
	//
	// Experimental.
	InputEnabled *bool `field:"optional" json:"inputEnabled" yaml:"inputEnabled"`
	// The action to take when PII is detected in the output.
	// Default: GuardrailAction.BLOCK
	//
	// Experimental.
	OutputAction GuardrailAction `field:"optional" json:"outputAction" yaml:"outputAction"`
	// Whether the PII filter is enabled for output.
	// Default: true.
	//
	// Experimental.
	OutputEnabled *bool `field:"optional" json:"outputEnabled" yaml:"outputEnabled"`
}

Interface to define a PII Filter.

Example: 

guardrail := bedrock.NewGuardrail(this, jsii.String("bedrockGuardrails"), &GuardrailProps{
	GuardrailName: jsii.String("my-BedrockGuardrails"),
})

// Add PII filter for addresses with input/output actions
guardrail.AddPIIFilter(&PIIFilter{
	Type: bedrock.GeneralPIIType_ADDRESS(),
	Action: bedrock.GuardrailAction_BLOCK,
	// below props are optional
	InputAction: bedrock.GuardrailAction_BLOCK,
	InputEnabled: jsii.Boolean(true),
	OutputAction: bedrock.GuardrailAction_ANONYMIZE,
	OutputEnabled: jsii.Boolean(true),
})

// Add PII filter for credit card numbers with input/output actions
guardrail.AddPIIFilter(&PIIFilter{
	Type: bedrock.FinancePIIType_CREDIT_DEBIT_CARD_NUMBER(),
	Action: bedrock.GuardrailAction_BLOCK,
	// below props are optional
	InputAction: bedrock.GuardrailAction_BLOCK,
	InputEnabled: jsii.Boolean(true),
	OutputAction: bedrock.GuardrailAction_ANONYMIZE,
	OutputEnabled: jsii.Boolean(true),
})

// Add PII filter for email addresses
guardrail.AddPIIFilter(&PIIFilter{
	Type: bedrock.GeneralPIIType_EMAIL(),
	Action: bedrock.GuardrailAction_ANONYMIZE,
})

// Add PII filter for US Social Security Numbers
guardrail.AddPIIFilter(&PIIFilter{
	Type: bedrock.USASpecificPIIType_US_SOCIAL_SECURITY_NUMBER(),
	Action: bedrock.GuardrailAction_BLOCK,
})

// Add PII filter for IP addresses
guardrail.AddPIIFilter(&PIIFilter{
	Type: bedrock.InformationTechnologyPIIType_IP_ADDRESS(),
	Action: bedrock.GuardrailAction_ANONYMIZE,
})

Experimental.

type PIIType 

type PIIType interface {
	// The string value of the PII type.
	// Experimental.
	Value() *string
	// Returns the string representation of the PII type.
	//
	// Returns: The string value of the PII type.
	// Experimental.
	ToString() *string
}

Abstract base class for all PII types.

Example: 

guardrail := bedrock.NewGuardrail(this, jsii.String("bedrockGuardrails"), &GuardrailProps{
	GuardrailName: jsii.String("my-BedrockGuardrails"),
})

// Add PII filter for addresses with input/output actions
guardrail.AddPIIFilter(&PIIFilter{
	Type: bedrock.GeneralPIIType_ADDRESS(),
	Action: bedrock.GuardrailAction_BLOCK,
	// below props are optional
	InputAction: bedrock.GuardrailAction_BLOCK,
	InputEnabled: jsii.Boolean(true),
	OutputAction: bedrock.GuardrailAction_ANONYMIZE,
	OutputEnabled: jsii.Boolean(true),
})

// Add PII filter for credit card numbers with input/output actions
guardrail.AddPIIFilter(&PIIFilter{
	Type: bedrock.FinancePIIType_CREDIT_DEBIT_CARD_NUMBER(),
	Action: bedrock.GuardrailAction_BLOCK,
	// below props are optional
	InputAction: bedrock.GuardrailAction_BLOCK,
	InputEnabled: jsii.Boolean(true),
	OutputAction: bedrock.GuardrailAction_ANONYMIZE,
	OutputEnabled: jsii.Boolean(true),
})

// Add PII filter for email addresses
guardrail.AddPIIFilter(&PIIFilter{
	Type: bedrock.GeneralPIIType_EMAIL(),
	Action: bedrock.GuardrailAction_ANONYMIZE,
})

// Add PII filter for US Social Security Numbers
guardrail.AddPIIFilter(&PIIFilter{
	Type: bedrock.USASpecificPIIType_US_SOCIAL_SECURITY_NUMBER(),
	Action: bedrock.GuardrailAction_BLOCK,
})

// Add PII filter for IP addresses
guardrail.AddPIIFilter(&PIIFilter{
	Type: bedrock.InformationTechnologyPIIType_IP_ADDRESS(),
	Action: bedrock.GuardrailAction_ANONYMIZE,
})

Experimental.

type ParameterType 

type ParameterType string

Enum for parameter types in function schemas.

Example: 

actionGroupFunction := lambda.NewFunction(this, jsii.String("ActionGroupFunction"), &FunctionProps{
	Runtime: lambda.Runtime_PYTHON_3_12(),
	Handler: jsii.String("index.handler"),
	Code: lambda.Code_FromAsset(path.join(__dirname, jsii.String("../lambda/action-group"))),
})

// Define a function schema with parameters
functionSchema := bedrock.NewFunctionSchema(&FunctionSchemaProps{
	Functions: []FunctionProps{
		&FunctionProps{
			Name: jsii.String("searchBooks"),
			Description: jsii.String("Search for books in the library catalog"),
			Parameters: map[string]FunctionParameterProps{
				"query": &FunctionParameterProps{
					"type": bedrock.ParameterType_STRING,
					"required": jsii.Boolean(true),
					"description": jsii.String("The search query string"),
				},
				"maxResults": &FunctionParameterProps{
					"type": bedrock.ParameterType_INTEGER,
					"required": jsii.Boolean(false),
					"description": jsii.String("Maximum number of results to return"),
				},
				"includeOutOfPrint": &FunctionParameterProps{
					"type": bedrock.ParameterType_BOOLEAN,
					"required": jsii.Boolean(false),
					"description": jsii.String("Whether to include out-of-print books"),
				},
			},
			RequireConfirmation: bedrock.RequireConfirmation_DISABLED,
		},
		&FunctionProps{
			Name: jsii.String("getBookDetails"),
			Description: jsii.String("Get detailed information about a specific book"),
			Parameters: map[string]FunctionParameterProps{
				"bookId": &FunctionParameterProps{
					"type": bedrock.ParameterType_STRING,
					"required": jsii.Boolean(true),
					"description": jsii.String("The unique identifier of the book"),
				},
			},
			RequireConfirmation: bedrock.RequireConfirmation_ENABLED,
		},
	},
})

// Create an action group using the function schema
actionGroup := bedrock.NewAgentActionGroup(&AgentActionGroupProps{
	Name: jsii.String("library-functions"),
	Description: jsii.String("Functions for interacting with the library catalog"),
	Executor: bedrock.ActionGroupExecutor_FromLambda(actionGroupFunction),
	FunctionSchema: functionSchema,
	Enabled: jsii.Boolean(true),
})

agent := bedrock.NewAgent(this, jsii.String("Agent"), &AgentProps{
	FoundationModel: bedrock.BedrockFoundationModel_ANTHROPIC_CLAUDE_HAIKU_V1_0(),
	Instruction: jsii.String("You are a helpful and friendly agent that answers questions about literature."),
	ActionGroups: []AgentActionGroup{
		actionGroup,
	},
})

Experimental. 

const (
	// String parameter type.
	// Experimental.
	ParameterType_STRING ParameterType = "STRING"
	// Number parameter type.
	// Experimental.
	ParameterType_NUMBER ParameterType = "NUMBER"
	// Integer parameter type.
	// Experimental.
	ParameterType_INTEGER ParameterType = "INTEGER"
	// Boolean parameter type.
	// Experimental.
	ParameterType_BOOLEAN ParameterType = "BOOLEAN"
	// Array parameter type.
	// Experimental.
	ParameterType_ARRAY ParameterType = "ARRAY"
	// Object parameter type.
	// Experimental.
	ParameterType_OBJECT ParameterType = "OBJECT"
)

type ParentActionGroupSignature 

type ParentActionGroupSignature interface {
	// The AWS-defined signature value for this action group capability.
	// Experimental.
	Value() *string
	// Returns the string representation of the signature value.
	//
	// Used when configuring the action group in CloudFormation.
	// Experimental.
	ToString() *string
}

AWS Defined signatures for enabling certain capabilities in your agent.

Example: 

// The code below shows an example of how to instantiate this type.
// The values are placeholders you should change.
import bedrock_alpha "github.com/aws/aws-cdk-go/awsbedrockalpha"

parentActionGroupSignature := bedrock_alpha.NewParentActionGroupSignature(jsii.String("value"))

Experimental.

func NewParentActionGroupSignature 

func NewParentActionGroupSignature(value *string) ParentActionGroupSignature

Constructor should be used as a temporary solution when a new signature is supported but its implementation in CDK hasn't been added yet. Experimental.

func ParentActionGroupSignature_CODE_INTERPRETER 

func ParentActionGroupSignature_CODE_INTERPRETER() ParentActionGroupSignature

func ParentActionGroupSignature_USER_INPUT 

func ParentActionGroupSignature_USER_INPUT() ParentActionGroupSignature

type Prompt 

type Prompt interface {
	PromptBase
	IPrompt
	// The description of the prompt.
	// Experimental.
	Description() *string
	// The environment this resource belongs to.
	//
	// For resources that are created and managed in a Stack (those created by
	// creating new class instances like `new Role()`, `new Bucket()`, etc.), this
	// is always the same as the environment of the stack they belong to.
	//
	// For referenced resources (those obtained from referencing methods like
	// `Role.fromRoleArn()`, `Bucket.fromBucketName()`, etc.), they might be
	// different than the stack they were imported into.
	// Experimental.
	Env() *interfaces.ResourceEnvironment
	// The KMS key that the prompt is encrypted with.
	// Experimental.
	KmsKey() awskms.IKey
	// The tree node.
	// Experimental.
	Node() constructs.Node
	// Returns a string-encoded token that resolves to the physical name that should be passed to the CloudFormation resource.
	//
	// This value will resolve to one of the following:
	// - a concrete value (e.g. `"my-awesome-bucket"`)
	// - `undefined`, when a name should be generated by CloudFormation
	// - a concrete name generated automatically during synthesis, in
	//   cross-environment scenarios.
	// Experimental.
	PhysicalName() *string
	// The ARN of the prompt.
	//
	// Example:
	//   "arn:aws:bedrock:us-east-1:123456789012:prompt/PROMPT12345"
	//
	// Experimental.
	PromptArn() *string
	// The ID of the prompt.
	//
	// Example:
	//   "PROMPT12345"
	//
	// Experimental.
	PromptId() *string
	// The name of the prompt.
	// Experimental.
	PromptName() *string
	// The version of the prompt.
	// Experimental.
	PromptVersion() *string
	// The stack in which this resource is defined.
	// Experimental.
	Stack() awscdk.Stack
	// The variants of the prompt.
	// Experimental.
	Variants() *[]IPromptVariant
	// Adds a prompt variant to the prompt.
	// Experimental.
	AddVariant(variant IPromptVariant)
	// Apply the given removal policy to this resource.
	//
	// The Removal Policy controls what happens to this resource when it stops
	// being managed by CloudFormation, either because you've removed it from the
	// CDK application or because you've made a change that requires the resource
	// to be replaced.
	//
	// The resource can be deleted (`RemovalPolicy.DESTROY`), or left in your AWS
	// account for data recovery and cleanup later (`RemovalPolicy.RETAIN`).
	// Experimental.
	ApplyRemovalPolicy(policy awscdk.RemovalPolicy)
	// Creates a prompt version, a static snapshot of your prompt that can be deployed to production.
	//
	// Returns: A PromptVersion object containing the version details including ARN and version string.
	// Default: - No description provided.
	//
	// Experimental.
	CreateVersion(description *string) PromptVersion
	// Experimental.
	GeneratePhysicalName() *string
	// Returns an environment-sensitive token that should be used for the resource's "ARN" attribute (e.g. `bucket.bucketArn`).
	//
	// Normally, this token will resolve to `arnAttr`, but if the resource is
	// referenced across environments, `arnComponents` will be used to synthesize
	// a concrete ARN with the resource's physical name. Make sure to reference
	// `this.physicalName` in `arnComponents`.
	// Experimental.
	GetResourceArnAttribute(arnAttr *string, arnComponents *awscdk.ArnComponents) *string
	// Returns an environment-sensitive token that should be used for the resource's "name" attribute (e.g. `bucket.bucketName`).
	//
	// Normally, this token will resolve to `nameAttr`, but if the resource is
	// referenced across environments, it will be resolved to `this.physicalName`,
	// which will be a concrete name.
	// Experimental.
	GetResourceNameAttribute(nameAttr *string) *string
	// Grant the given identity permissions to get the prompt.
	//
	// [disable-awslint:no-grants].
	//
	// Returns: An IAM Grant object representing the granted permissions.
	// Default: - Default grant configuration:
	// - actions: ['bedrock:GetPrompt']
	// - resourceArns: [this.promptArn]
	//
	// Experimental.
	GrantGet(grantee awsiam.IGrantable) awsiam.Grant
	// Returns a string representation of this construct.
	// Experimental.
	ToString() *string
	// Applies one or more mixins to this construct.
	//
	// Mixins are applied in order. The list of constructs is captured at the
	// start of the call, so constructs added by a mixin will not be visited.
	// Use multiple `with()` calls if subsequent mixins should apply to added
	// constructs.
	// Experimental.
	With(mixins ...constructs.IMixin) constructs.IConstruct
}

Class to create (or import) a Prompt with CDK.

Prompts are a specific set of inputs that guide Foundation Models (FMs) on Amazon Bedrock to generate an appropriate response or output for a given task or instruction. You can optimize the prompt for specific use cases and models.

Example: 

cmk := kms.NewKey(this, jsii.String("cmk"), &KeyProps{
})
claudeModel := bedrock.BedrockFoundationModel_ANTHROPIC_CLAUDE_SONNET_V1_0()

variant1 := bedrock.PromptVariant_Text(&TextPromptVariantProps{
	VariantName: jsii.String("variant1"),
	Model: claudeModel,
	PromptVariables: []*string{
		jsii.String("topic"),
	},
	PromptText: jsii.String("This is my first text prompt. Please summarize our conversation on: {{topic}}."),
	InferenceConfiguration: bedrock.PromptInferenceConfiguration_Text(&PromptInferenceConfigurationProps{
		Temperature: jsii.Number(1),
		TopP: jsii.Number(0.999),
		MaxTokens: jsii.Number(2000),
	}),
})

prompt1 := bedrock.NewPrompt(this, jsii.String("prompt1"), &PromptProps{
	PromptName: jsii.String("prompt1"),
	Description: jsii.String("my first prompt"),
	DefaultVariant: variant1,
	Variants: []IPromptVariant{
		variant1,
	},
	KmsKey: cmk,
})

See: https://docs.aws.amazon.com/bedrock/latest/userguide/prompt-management.html

Experimental.

func NewPrompt 

func NewPrompt(scope constructs.Construct, id *string, props *PromptProps) Prompt

**************************************************************************** CONSTRUCTOR ***************************************************************************. Experimental.

type PromptAttributes 

type PromptAttributes struct {
	// The ARN of the prompt.
	//
	// Example:
	//   "arn:aws:bedrock:us-east-1:123456789012:prompt/PROMPT12345"
	//
	// Experimental.
	PromptArn *string `field:"required" json:"promptArn" yaml:"promptArn"`
	// Optional KMS encryption key associated with this prompt.
	// Default: undefined - An AWS managed key is used.
	//
	// Experimental.
	KmsKey awskms.IKey `field:"optional" json:"kmsKey" yaml:"kmsKey"`
	// The version of the prompt.
	// Default: "DRAFT".
	//
	// Experimental.
	PromptVersion *string `field:"optional" json:"promptVersion" yaml:"promptVersion"`
}

Attributes for specifying an imported Bedrock Prompt.

Example: 

// Import an existing prompt by ARN
importedPrompt := bedrock.Prompt_FromPromptAttributes(this, jsii.String("ImportedPrompt"), &PromptAttributes{
	PromptArn: jsii.String("arn:aws:bedrock:region:account:prompt/prompt-id"),
	KmsKey: kms.Key_FromKeyArn(this, jsii.String("ImportedKey"), jsii.String("arn:aws:kms:region:account:key/key-id")),
	 // optional
	PromptVersion: jsii.String("1"),
})

Experimental.

type PromptBase 

type PromptBase interface {
	awscdk.Resource
	IPrompt
	// The environment this resource belongs to.
	//
	// For resources that are created and managed in a Stack (those created by
	// creating new class instances like `new Role()`, `new Bucket()`, etc.), this
	// is always the same as the environment of the stack they belong to.
	//
	// For referenced resources (those obtained from referencing methods like
	// `Role.fromRoleArn()`, `Bucket.fromBucketName()`, etc.), they might be
	// different than the stack they were imported into.
	// Experimental.
	Env() *interfaces.ResourceEnvironment
	// Optional KMS encryption key associated with this prompt.
	// Experimental.
	KmsKey() awskms.IKey
	// The tree node.
	// Experimental.
	Node() constructs.Node
	// Returns a string-encoded token that resolves to the physical name that should be passed to the CloudFormation resource.
	//
	// This value will resolve to one of the following:
	// - a concrete value (e.g. `"my-awesome-bucket"`)
	// - `undefined`, when a name should be generated by CloudFormation
	// - a concrete name generated automatically during synthesis, in
	//   cross-environment scenarios.
	// Experimental.
	PhysicalName() *string
	// The ARN of the prompt.
	// Experimental.
	PromptArn() *string
	// The ID of the prompt.
	// Experimental.
	PromptId() *string
	// The version of the prompt.
	// Experimental.
	PromptVersion() *string
	// The stack in which this resource is defined.
	// Experimental.
	Stack() awscdk.Stack
	// Apply the given removal policy to this resource.
	//
	// The Removal Policy controls what happens to this resource when it stops
	// being managed by CloudFormation, either because you've removed it from the
	// CDK application or because you've made a change that requires the resource
	// to be replaced.
	//
	// The resource can be deleted (`RemovalPolicy.DESTROY`), or left in your AWS
	// account for data recovery and cleanup later (`RemovalPolicy.RETAIN`).
	// Experimental.
	ApplyRemovalPolicy(policy awscdk.RemovalPolicy)
	// Experimental.
	GeneratePhysicalName() *string
	// Returns an environment-sensitive token that should be used for the resource's "ARN" attribute (e.g. `bucket.bucketArn`).
	//
	// Normally, this token will resolve to `arnAttr`, but if the resource is
	// referenced across environments, `arnComponents` will be used to synthesize
	// a concrete ARN with the resource's physical name. Make sure to reference
	// `this.physicalName` in `arnComponents`.
	// Experimental.
	GetResourceArnAttribute(arnAttr *string, arnComponents *awscdk.ArnComponents) *string
	// Returns an environment-sensitive token that should be used for the resource's "name" attribute (e.g. `bucket.bucketName`).
	//
	// Normally, this token will resolve to `nameAttr`, but if the resource is
	// referenced across environments, it will be resolved to `this.physicalName`,
	// which will be a concrete name.
	// Experimental.
	GetResourceNameAttribute(nameAttr *string) *string
	// Grant the given identity permissions to get the prompt.
	//
	// [disable-awslint:no-grants].
	//
	// Returns: An IAM Grant object representing the granted permissions.
	// Default: - Default grant configuration:
	// - actions: ['bedrock:GetPrompt']
	// - resourceArns: [this.promptArn]
	//
	// Experimental.
	GrantGet(grantee awsiam.IGrantable) awsiam.Grant
	// Returns a string representation of this construct.
	// Experimental.
	ToString() *string
	// Applies one or more mixins to this construct.
	//
	// Mixins are applied in order. The list of constructs is captured at the
	// start of the call, so constructs added by a mixin will not be visited.
	// Use multiple `with()` calls if subsequent mixins should apply to added
	// constructs.
	// Experimental.
	With(mixins ...constructs.IMixin) constructs.IConstruct
}

Abstract base class for a Prompt.

Contains methods and attributes valid for Prompts either created with CDK or imported. Experimental.

type PromptGenAiResource 

type PromptGenAiResource interface {
}

Abstract base class for prompt GenAI resource configurations.

This provides a high-level abstraction over the underlying CloudFormation GenAI resource properties, offering a more developer-friendly interface while maintaining full compatibility with the underlying AWS Bedrock service.

Example: 

// The code below shows an example of how to instantiate this type.
// The values are placeholders you should change.
import bedrock_alpha "github.com/aws/aws-cdk-go/awsbedrockalpha"

var agentAlias AgentAlias

promptGenAiResource := bedrock_alpha.PromptGenAiResource_Agent(&AgentGenAiResourceProps{
	AgentAlias: agentAlias,
})

Experimental.

func PromptGenAiResource_Agent 

func PromptGenAiResource_Agent(props *AgentGenAiResourceProps) PromptGenAiResource

Creates an agent GenAI resource configuration. Experimental.

type PromptInferenceConfiguration 

type PromptInferenceConfiguration interface {
}

Abstract base class for prompt inference configurations.

This provides a high-level abstraction over the underlying CloudFormation inference configuration properties, offering a more developer-friendly interface while maintaining full compatibility with the underlying AWS Bedrock service.

Example: 

cmk := kms.NewKey(this, jsii.String("cmk"), &KeyProps{
})
claudeModel := bedrock.BedrockFoundationModel_ANTHROPIC_CLAUDE_SONNET_V1_0()

variant1 := bedrock.PromptVariant_Text(&TextPromptVariantProps{
	VariantName: jsii.String("variant1"),
	Model: claudeModel,
	PromptVariables: []*string{
		jsii.String("topic"),
	},
	PromptText: jsii.String("This is my first text prompt. Please summarize our conversation on: {{topic}}."),
	InferenceConfiguration: bedrock.PromptInferenceConfiguration_Text(&PromptInferenceConfigurationProps{
		Temperature: jsii.Number(1),
		TopP: jsii.Number(0.999),
		MaxTokens: jsii.Number(2000),
	}),
})

prompt1 := bedrock.NewPrompt(this, jsii.String("prompt1"), &PromptProps{
	PromptName: jsii.String("prompt1"),
	Description: jsii.String("my first prompt"),
	DefaultVariant: variant1,
	Variants: []IPromptVariant{
		variant1,
	},
	KmsKey: cmk,
})

Experimental.

func PromptInferenceConfiguration_Text 

func PromptInferenceConfiguration_Text(props *PromptInferenceConfigurationProps) PromptInferenceConfiguration

Creates a text inference configuration. Experimental.

type PromptInferenceConfigurationProps 

type PromptInferenceConfigurationProps struct {
	// The maximum number of tokens to return in the response.
	// Default: - No limit specified.
	//
	// Experimental.
	MaxTokens *float64 `field:"optional" json:"maxTokens" yaml:"maxTokens"`
	// A list of strings that define sequences after which the model will stop generating.
	// Default: - No stop sequences.
	//
	// Experimental.
	StopSequences *[]*string `field:"optional" json:"stopSequences" yaml:"stopSequences"`
	// Controls the randomness of the response.
	//
	// Higher values make output more random, lower values more deterministic.
	// Valid range is 0.0 to 1.0.
	// Default: - Model default temperature.
	//
	// Experimental.
	Temperature *float64 `field:"optional" json:"temperature" yaml:"temperature"`
	// The percentage of most-likely candidates that the model considers for the next token.
	//
	// Valid range is 0.0 to 1.0.
	// Default: - Model default topP.
	//
	// Experimental.
	TopP *float64 `field:"optional" json:"topP" yaml:"topP"`
}

Properties for creating a prompt inference configuration.

Example: 

cmk := kms.NewKey(this, jsii.String("cmk"), &KeyProps{
})
claudeModel := bedrock.BedrockFoundationModel_ANTHROPIC_CLAUDE_SONNET_V1_0()

variant1 := bedrock.PromptVariant_Text(&TextPromptVariantProps{
	VariantName: jsii.String("variant1"),
	Model: claudeModel,
	PromptVariables: []*string{
		jsii.String("topic"),
	},
	PromptText: jsii.String("This is my first text prompt. Please summarize our conversation on: {{topic}}."),
	InferenceConfiguration: bedrock.PromptInferenceConfiguration_Text(&PromptInferenceConfigurationProps{
		Temperature: jsii.Number(1),
		TopP: jsii.Number(0.999),
		MaxTokens: jsii.Number(2000),
	}),
})

prompt1 := bedrock.NewPrompt(this, jsii.String("prompt1"), &PromptProps{
	PromptName: jsii.String("prompt1"),
	Description: jsii.String("my first prompt"),
	DefaultVariant: variant1,
	Variants: []IPromptVariant{
		variant1,
	},
	KmsKey: cmk,
})

Experimental.

type PromptKnowledgeBaseResponseGenerationConfigCustomParser 

type PromptKnowledgeBaseResponseGenerationConfigCustomParser struct {
	// The type of step this configuration applies to.
	// Experimental.
	StepType AgentStepType `field:"required" json:"stepType" yaml:"stepType"`
	// The custom prompt template to be used.
	// See: https://docs.aws.amazon.com/bedrock/latest/userguide/prompt-placeholders.html
	//
	// Default: - The default prompt template will be used.
	//
	// Experimental.
	CustomPromptTemplate *string `field:"optional" json:"customPromptTemplate" yaml:"customPromptTemplate"`
	// The inference configuration parameters to use.
	// Default: undefined - Default inference configuration will be used.
	//
	// Experimental.
	InferenceConfig *InferenceConfiguration `field:"optional" json:"inferenceConfig" yaml:"inferenceConfig"`
	// Whether to enable or skip this step in the agent sequence.
	// Default: - The default state for each step type is as follows.
	//
	// PRE_PROCESSING – ENABLED
	// ORCHESTRATION – ENABLED
	// KNOWLEDGE_BASE_RESPONSE_GENERATION – ENABLED
	// POST_PROCESSING – DISABLED.
	//
	// Experimental.
	StepEnabled *bool `field:"optional" json:"stepEnabled" yaml:"stepEnabled"`
	// Whether to use the custom Lambda parser defined for the sequence.
	// Default: - false.
	//
	// Experimental.
	UseCustomParser *bool `field:"optional" json:"useCustomParser" yaml:"useCustomParser"`
}

Configuration for the knowledge base response generation step.

Example: 

// The code below shows an example of how to instantiate this type.
// The values are placeholders you should change.
import bedrock_alpha "github.com/aws/aws-cdk-go/awsbedrockalpha"

promptKnowledgeBaseResponseGenerationConfigCustomParser := &PromptKnowledgeBaseResponseGenerationConfigCustomParser{
	StepType: bedrock_alpha.AgentStepType_PRE_PROCESSING,

	// the properties below are optional
	CustomPromptTemplate: jsii.String("customPromptTemplate"),
	InferenceConfig: &InferenceConfiguration{
		MaximumLength: jsii.Number(123),
		StopSequences: []*string{
			jsii.String("stopSequences"),
		},
		Temperature: jsii.Number(123),
		TopK: jsii.Number(123),
		TopP: jsii.Number(123),
	},
	StepEnabled: jsii.Boolean(false),
	UseCustomParser: jsii.Boolean(false),
}

Experimental.

type PromptMemorySummarizationConfigCustomParser 

type PromptMemorySummarizationConfigCustomParser struct {
	// The type of step this configuration applies to.
	// Experimental.
	StepType AgentStepType `field:"required" json:"stepType" yaml:"stepType"`
	// The custom prompt template to be used.
	// See: https://docs.aws.amazon.com/bedrock/latest/userguide/prompt-placeholders.html
	//
	// Default: - The default prompt template will be used.
	//
	// Experimental.
	CustomPromptTemplate *string `field:"optional" json:"customPromptTemplate" yaml:"customPromptTemplate"`
	// The inference configuration parameters to use.
	// Default: undefined - Default inference configuration will be used.
	//
	// Experimental.
	InferenceConfig *InferenceConfiguration `field:"optional" json:"inferenceConfig" yaml:"inferenceConfig"`
	// Whether to enable or skip this step in the agent sequence.
	// Default: - The default state for each step type is as follows.
	//
	// PRE_PROCESSING – ENABLED
	// ORCHESTRATION – ENABLED
	// KNOWLEDGE_BASE_RESPONSE_GENERATION – ENABLED
	// POST_PROCESSING – DISABLED.
	//
	// Experimental.
	StepEnabled *bool `field:"optional" json:"stepEnabled" yaml:"stepEnabled"`
	// Whether to use the custom Lambda parser defined for the sequence.
	// Default: - false.
	//
	// Experimental.
	UseCustomParser *bool `field:"optional" json:"useCustomParser" yaml:"useCustomParser"`
}

Configuration for the memory summarization step.

Example: 

// The code below shows an example of how to instantiate this type.
// The values are placeholders you should change.
import bedrock_alpha "github.com/aws/aws-cdk-go/awsbedrockalpha"

promptMemorySummarizationConfigCustomParser := &PromptMemorySummarizationConfigCustomParser{
	StepType: bedrock_alpha.AgentStepType_PRE_PROCESSING,

	// the properties below are optional
	CustomPromptTemplate: jsii.String("customPromptTemplate"),
	InferenceConfig: &InferenceConfiguration{
		MaximumLength: jsii.Number(123),
		StopSequences: []*string{
			jsii.String("stopSequences"),
		},
		Temperature: jsii.Number(123),
		TopK: jsii.Number(123),
		TopP: jsii.Number(123),
	},
	StepEnabled: jsii.Boolean(false),
	UseCustomParser: jsii.Boolean(false),
}

Experimental.

type PromptOrchestrationConfigCustomParser 

type PromptOrchestrationConfigCustomParser struct {
	// The type of step this configuration applies to.
	// Experimental.
	StepType AgentStepType `field:"required" json:"stepType" yaml:"stepType"`
	// The custom prompt template to be used.
	// See: https://docs.aws.amazon.com/bedrock/latest/userguide/prompt-placeholders.html
	//
	// Default: - The default prompt template will be used.
	//
	// Experimental.
	CustomPromptTemplate *string `field:"optional" json:"customPromptTemplate" yaml:"customPromptTemplate"`
	// The inference configuration parameters to use.
	// Default: undefined - Default inference configuration will be used.
	//
	// Experimental.
	InferenceConfig *InferenceConfiguration `field:"optional" json:"inferenceConfig" yaml:"inferenceConfig"`
	// Whether to enable or skip this step in the agent sequence.
	// Default: - The default state for each step type is as follows.
	//
	// PRE_PROCESSING – ENABLED
	// ORCHESTRATION – ENABLED
	// KNOWLEDGE_BASE_RESPONSE_GENERATION – ENABLED
	// POST_PROCESSING – DISABLED.
	//
	// Experimental.
	StepEnabled *bool `field:"optional" json:"stepEnabled" yaml:"stepEnabled"`
	// Whether to use the custom Lambda parser defined for the sequence.
	// Default: - false.
	//
	// Experimental.
	UseCustomParser *bool `field:"optional" json:"useCustomParser" yaml:"useCustomParser"`
}

Configuration for the orchestration step.

Example: 

// The code below shows an example of how to instantiate this type.
// The values are placeholders you should change.
import bedrock_alpha "github.com/aws/aws-cdk-go/awsbedrockalpha"

promptOrchestrationConfigCustomParser := &PromptOrchestrationConfigCustomParser{
	StepType: bedrock_alpha.AgentStepType_PRE_PROCESSING,

	// the properties below are optional
	CustomPromptTemplate: jsii.String("customPromptTemplate"),
	InferenceConfig: &InferenceConfiguration{
		MaximumLength: jsii.Number(123),
		StopSequences: []*string{
			jsii.String("stopSequences"),
		},
		Temperature: jsii.Number(123),
		TopK: jsii.Number(123),
		TopP: jsii.Number(123),
	},
	StepEnabled: jsii.Boolean(false),
	UseCustomParser: jsii.Boolean(false),
}

Experimental.

type PromptOverrideConfiguration 

type PromptOverrideConfiguration interface {
	// Configuration for the knowledge base response generation step.
	// Experimental.
	KnowledgeBaseResponseGenerationStep() *PromptKnowledgeBaseResponseGenerationConfigCustomParser
	// Configuration for the memory summarization step.
	// Experimental.
	MemorySummarizationStep() *PromptMemorySummarizationConfigCustomParser
	// Configuration for the orchestration step.
	// Experimental.
	OrchestrationStep() *PromptOrchestrationConfigCustomParser
	// The custom Lambda parser function to use.
	//
	// The Lambda parser processes and interprets the raw foundation model output.
	// It receives an input event with:
	// - messageVersion: Version of message format (1.0)
	// - agent: Info about the agent (name, id, alias, version)
	// - invokeModelRawResponse: Raw model output to parse
	// - promptType: Type of prompt being parsed
	// - overrideType: Type of override (OUTPUT_PARSER)
	//
	// The Lambda must return a response that the agent uses for next actions.
	// See: https://docs.aws.amazon.com/bedrock/latest/userguide/lambda-parser.html
	//
	// Experimental.
	Parser() awslambda.IFunction
	// Configuration for the post-processing step.
	// Experimental.
	PostProcessingStep() *PromptPostProcessingConfigCustomParser
	// Configuration for the pre-processing step.
	// Experimental.
	PreProcessingStep() *PromptPreProcessingConfigCustomParser
	// Configuration for the routing classifier step.
	// Experimental.
	RoutingClassifierStep() *PromptRoutingClassifierConfigCustomParser
}

Configuration for overriding prompt templates and behaviors in different parts of an agent's sequence.

This allows customizing how the agent processes inputs, makes decisions, and generates responses.

Example: 

agent := bedrock.NewAgent(this, jsii.String("Agent"), &AgentProps{
	FoundationModel: bedrock.BedrockFoundationModel_AMAZON_NOVA_LITE_V1(),
	Instruction: jsii.String("You are a helpful assistant."),
	PromptOverrideConfiguration: bedrock.PromptOverrideConfiguration_FromSteps([]PromptStepConfigBase{
		&PromptRoutingClassifierConfigCustomParser{
			StepType: bedrock.AgentStepType_ROUTING_CLASSIFIER,
			StepEnabled: jsii.Boolean(true),
			CustomPromptTemplate: jsii.String("Your routing template here"),
			FoundationModel: bedrock.BedrockFoundationModel_ANTHROPIC_CLAUDE_V2(),
		}.(PromptRoutingClassifierConfigCustomParser),
	}),
})

Experimental.

func PromptOverrideConfiguration_FromSteps 

func PromptOverrideConfiguration_FromSteps(steps *[]*PromptStepConfigBase) PromptOverrideConfiguration

Creates a PromptOverrideConfiguration from individual step configurations.

Use this method when you want to override prompts without using a custom parser.

Returns: A new PromptOverrideConfiguration instance. Experimental.

func PromptOverrideConfiguration_WithCustomParser 

func PromptOverrideConfiguration_WithCustomParser(props *CustomParserProps) PromptOverrideConfiguration

Creates a PromptOverrideConfiguration with a custom Lambda parser function. Experimental.

type PromptPostProcessingConfigCustomParser 

type PromptPostProcessingConfigCustomParser struct {
	// The type of step this configuration applies to.
	// Experimental.
	StepType AgentStepType `field:"required" json:"stepType" yaml:"stepType"`
	// The custom prompt template to be used.
	// See: https://docs.aws.amazon.com/bedrock/latest/userguide/prompt-placeholders.html
	//
	// Default: - The default prompt template will be used.
	//
	// Experimental.
	CustomPromptTemplate *string `field:"optional" json:"customPromptTemplate" yaml:"customPromptTemplate"`
	// The inference configuration parameters to use.
	// Default: undefined - Default inference configuration will be used.
	//
	// Experimental.
	InferenceConfig *InferenceConfiguration `field:"optional" json:"inferenceConfig" yaml:"inferenceConfig"`
	// Whether to enable or skip this step in the agent sequence.
	// Default: - The default state for each step type is as follows.
	//
	// PRE_PROCESSING – ENABLED
	// ORCHESTRATION – ENABLED
	// KNOWLEDGE_BASE_RESPONSE_GENERATION – ENABLED
	// POST_PROCESSING – DISABLED.
	//
	// Experimental.
	StepEnabled *bool `field:"optional" json:"stepEnabled" yaml:"stepEnabled"`
	// Whether to use the custom Lambda parser defined for the sequence.
	// Default: - false.
	//
	// Experimental.
	UseCustomParser *bool `field:"optional" json:"useCustomParser" yaml:"useCustomParser"`
}

Configuration for the post-processing step.

Example: 

// The code below shows an example of how to instantiate this type.
// The values are placeholders you should change.
import bedrock_alpha "github.com/aws/aws-cdk-go/awsbedrockalpha"

promptPostProcessingConfigCustomParser := &PromptPostProcessingConfigCustomParser{
	StepType: bedrock_alpha.AgentStepType_PRE_PROCESSING,

	// the properties below are optional
	CustomPromptTemplate: jsii.String("customPromptTemplate"),
	InferenceConfig: &InferenceConfiguration{
		MaximumLength: jsii.Number(123),
		StopSequences: []*string{
			jsii.String("stopSequences"),
		},
		Temperature: jsii.Number(123),
		TopK: jsii.Number(123),
		TopP: jsii.Number(123),
	},
	StepEnabled: jsii.Boolean(false),
	UseCustomParser: jsii.Boolean(false),
}

Experimental.

type PromptPreProcessingConfigCustomParser 

type PromptPreProcessingConfigCustomParser struct {
	// The type of step this configuration applies to.
	// Experimental.
	StepType AgentStepType `field:"required" json:"stepType" yaml:"stepType"`
	// The custom prompt template to be used.
	// See: https://docs.aws.amazon.com/bedrock/latest/userguide/prompt-placeholders.html
	//
	// Default: - The default prompt template will be used.
	//
	// Experimental.
	CustomPromptTemplate *string `field:"optional" json:"customPromptTemplate" yaml:"customPromptTemplate"`
	// The inference configuration parameters to use.
	// Default: undefined - Default inference configuration will be used.
	//
	// Experimental.
	InferenceConfig *InferenceConfiguration `field:"optional" json:"inferenceConfig" yaml:"inferenceConfig"`
	// Whether to enable or skip this step in the agent sequence.
	// Default: - The default state for each step type is as follows.
	//
	// PRE_PROCESSING – ENABLED
	// ORCHESTRATION – ENABLED
	// KNOWLEDGE_BASE_RESPONSE_GENERATION – ENABLED
	// POST_PROCESSING – DISABLED.
	//
	// Experimental.
	StepEnabled *bool `field:"optional" json:"stepEnabled" yaml:"stepEnabled"`
	// Whether to use the custom Lambda parser defined for the sequence.
	// Default: - false.
	//
	// Experimental.
	UseCustomParser *bool `field:"optional" json:"useCustomParser" yaml:"useCustomParser"`
}

Configuration for the pre-processing step.

Example: 

parserFunction := lambda.NewFunction(this, jsii.String("ParserFunction"), &FunctionProps{
	Runtime: lambda.Runtime_PYTHON_3_10(),
	Handler: jsii.String("index.handler"),
	Code: lambda.Code_FromAsset(jsii.String("lambda")),
})

agent := bedrock.NewAgent(this, jsii.String("Agent"), &AgentProps{
	FoundationModel: bedrock.BedrockFoundationModel_AMAZON_NOVA_LITE_V1(),
	Instruction: jsii.String("You are a helpful assistant."),
	PromptOverrideConfiguration: bedrock.PromptOverrideConfiguration_WithCustomParser(&CustomParserProps{
		Parser: parserFunction,
		PreProcessingStep: &PromptPreProcessingConfigCustomParser{
			StepType: bedrock.AgentStepType_PRE_PROCESSING,
			UseCustomParser: jsii.Boolean(true),
		},
	}),
})

Experimental.

type PromptProps 

type PromptProps struct {
	// The name of the prompt.
	//
	// This will be used as the physical name of the prompt.
	// Allowed Pattern: ^([0-9a-zA-Z][_-]?){1,100}$
	// Experimental.
	PromptName *string `field:"required" json:"promptName" yaml:"promptName"`
	// The Prompt Variant that will be used by default.
	// Default: - No default variant provided.
	//
	// Experimental.
	DefaultVariant IPromptVariant `field:"optional" json:"defaultVariant" yaml:"defaultVariant"`
	// A description of what the prompt does.
	// Default: - No description provided.
	// Maximum Length: 200.
	//
	// Experimental.
	Description *string `field:"optional" json:"description" yaml:"description"`
	// The KMS key that the prompt is encrypted with.
	// Default: - AWS owned and managed key.
	//
	// Experimental.
	KmsKey awskms.IKey `field:"optional" json:"kmsKey" yaml:"kmsKey"`
	// Tags to apply to the prompt.
	// Default: - No tags applied.
	//
	// Experimental.
	Tags *map[string]*string `field:"optional" json:"tags" yaml:"tags"`
	// The variants of your prompt.
	//
	// Variants can use different messages, models,
	// or configurations so that you can compare their outputs to decide the best
	// variant for your use case. Maximum of 1 variants.
	// Default: - No additional variants provided.
	//
	// Experimental.
	Variants *[]IPromptVariant `field:"optional" json:"variants" yaml:"variants"`
}

Properties for creating a CDK managed Bedrock Prompt.

Example: 

cmk := kms.NewKey(this, jsii.String("cmk"), &KeyProps{
})
claudeModel := bedrock.BedrockFoundationModel_ANTHROPIC_CLAUDE_SONNET_V1_0()

variant1 := bedrock.PromptVariant_Text(&TextPromptVariantProps{
	VariantName: jsii.String("variant1"),
	Model: claudeModel,
	PromptVariables: []*string{
		jsii.String("topic"),
	},
	PromptText: jsii.String("This is my first text prompt. Please summarize our conversation on: {{topic}}."),
	InferenceConfiguration: bedrock.PromptInferenceConfiguration_Text(&PromptInferenceConfigurationProps{
		Temperature: jsii.Number(1),
		TopP: jsii.Number(0.999),
		MaxTokens: jsii.Number(2000),
	}),
})

prompt1 := bedrock.NewPrompt(this, jsii.String("prompt1"), &PromptProps{
	PromptName: jsii.String("prompt1"),
	Description: jsii.String("my first prompt"),
	DefaultVariant: variant1,
	Variants: []IPromptVariant{
		variant1,
	},
	KmsKey: cmk,
})

Experimental.

type PromptRouter 

type PromptRouter interface {
	IBedrockInvokable
	IPromptRouter
	// The ARN used for invoking this prompt router.
	//
	// This equals to the promptRouterArn property, useful for implementing IBedrockInvokable interface.
	// Experimental.
	InvokableArn() *string
	// The ARN of the prompt router.
	// Experimental.
	PromptRouterArn() *string
	// The ID of the prompt router.
	// Experimental.
	PromptRouterId() *string
	// The inference endpoints (cross-region profiles) that this router will route to.
	//
	// These are created automatically based on the routing models and region.
	// Experimental.
	RoutingEndpoints() *[]IBedrockInvokable
	// Grants the necessary permissions to invoke this prompt router and all its routing endpoints.
	//
	// This method grants permissions to:
	// - Get prompt router details (bedrock:GetPromptRouter)
	// - Invoke models through the router (bedrock:InvokeModel)
	// - Use all underlying models and cross-region profiles
	// [disable-awslint:no-grants].
	//
	// Returns: An IAM Grant object representing the granted permissions.
	// Experimental.
	GrantInvoke(grantee awsiam.IGrantable) awsiam.Grant
}

Amazon Bedrock intelligent prompt routing provides a single serverless endpoint for efficiently routing requests between different foundational models within the same model family.

It can help you optimize for response quality and cost.

Intelligent prompt routing predicts the performance of each model for each request, and dynamically routes each request to the model that it predicts is most likely to give the desired response at the lowest cost.

Example: 

// Create a prompt router for intelligent model selection
promptRouter := bedrock.PromptRouter_FromDefaultId(bedrock.DefaultPromptRouterIdentifier_ANTHROPIC_CLAUDE_V1(), jsii.String("us-east-1"))

// Use the prompt router with a prompt variant
variant := bedrock.PromptVariant_Text(&TextPromptVariantProps{
	VariantName: jsii.String("variant1"),
	PromptText: jsii.String("What is the capital of France?"),
	Model: promptRouter,
})

bedrock.NewPrompt(this, jsii.String("Prompt"), &PromptProps{
	PromptName: jsii.String("prompt-router-test"),
	Variants: []IPromptVariant{
		variant,
	},
})

See: https://docs.aws.amazon.com/bedrock/latest/userguide/prompt-routing.html

Experimental.

func NewPromptRouter 

func NewPromptRouter(props *PromptRouterProps, region *string) PromptRouter

Experimental.

func PromptRouter_FromDefaultId 

func PromptRouter_FromDefaultId(defaultRouter DefaultPromptRouterIdentifier, region *string) PromptRouter

Creates a PromptRouter from a default router identifier.

Returns: A new PromptRouter instance configured with the default settings. Experimental.

type PromptRouterProps 

type PromptRouterProps struct {
	// Prompt Router ID that identifies the routing configuration.
	// Experimental.
	PromptRouterId *string `field:"required" json:"promptRouterId" yaml:"promptRouterId"`
	// The foundation models this router will route to.
	//
	// The router will intelligently select between these models based on the request.
	// Experimental.
	RoutingModels *[]BedrockFoundationModel `field:"required" json:"routingModels" yaml:"routingModels"`
}

Properties for configuring a Prompt Router.

Example: 

// The code below shows an example of how to instantiate this type.
// The values are placeholders you should change.
import bedrock_alpha "github.com/aws/aws-cdk-go/awsbedrockalpha"

var bedrockFoundationModel BedrockFoundationModel

promptRouterProps := &PromptRouterProps{
	PromptRouterId: jsii.String("promptRouterId"),
	RoutingModels: []BedrockFoundationModel{
		bedrockFoundationModel,
	},
}

Experimental.

type PromptRoutingClassifierConfigCustomParser 

type PromptRoutingClassifierConfigCustomParser struct {
	// The type of step this configuration applies to.
	// Experimental.
	StepType AgentStepType `field:"required" json:"stepType" yaml:"stepType"`
	// The custom prompt template to be used.
	// See: https://docs.aws.amazon.com/bedrock/latest/userguide/prompt-placeholders.html
	//
	// Default: - The default prompt template will be used.
	//
	// Experimental.
	CustomPromptTemplate *string `field:"optional" json:"customPromptTemplate" yaml:"customPromptTemplate"`
	// The inference configuration parameters to use.
	// Default: undefined - Default inference configuration will be used.
	//
	// Experimental.
	InferenceConfig *InferenceConfiguration `field:"optional" json:"inferenceConfig" yaml:"inferenceConfig"`
	// Whether to enable or skip this step in the agent sequence.
	// Default: - The default state for each step type is as follows.
	//
	// PRE_PROCESSING – ENABLED
	// ORCHESTRATION – ENABLED
	// KNOWLEDGE_BASE_RESPONSE_GENERATION – ENABLED
	// POST_PROCESSING – DISABLED.
	//
	// Experimental.
	StepEnabled *bool `field:"optional" json:"stepEnabled" yaml:"stepEnabled"`
	// Whether to use the custom Lambda parser defined for the sequence.
	// Default: - false.
	//
	// Experimental.
	UseCustomParser *bool `field:"optional" json:"useCustomParser" yaml:"useCustomParser"`
	// The foundation model to use for the routing classifier step.
	//
	// This is required for the routing classifier step.
	// Experimental.
	FoundationModel IBedrockInvokable `field:"required" json:"foundationModel" yaml:"foundationModel"`
}

Configuration for the routing classifier step.

Example: 

// The code below shows an example of how to instantiate this type.
// The values are placeholders you should change.
import bedrock_alpha "github.com/aws/aws-cdk-go/awsbedrockalpha"

var bedrockInvokable IBedrockInvokable

promptRoutingClassifierConfigCustomParser := &PromptRoutingClassifierConfigCustomParser{
	FoundationModel: bedrockInvokable,
	StepType: bedrock_alpha.AgentStepType_PRE_PROCESSING,

	// the properties below are optional
	CustomPromptTemplate: jsii.String("customPromptTemplate"),
	InferenceConfig: &InferenceConfiguration{
		MaximumLength: jsii.Number(123),
		StopSequences: []*string{
			jsii.String("stopSequences"),
		},
		Temperature: jsii.Number(123),
		TopK: jsii.Number(123),
		TopP: jsii.Number(123),
	},
	StepEnabled: jsii.Boolean(false),
	UseCustomParser: jsii.Boolean(false),
}

Experimental.

type PromptStepConfigBase 

type PromptStepConfigBase struct {
	// The type of step this configuration applies to.
	// Experimental.
	StepType AgentStepType `field:"required" json:"stepType" yaml:"stepType"`
	// The custom prompt template to be used.
	// See: https://docs.aws.amazon.com/bedrock/latest/userguide/prompt-placeholders.html
	//
	// Default: - The default prompt template will be used.
	//
	// Experimental.
	CustomPromptTemplate *string `field:"optional" json:"customPromptTemplate" yaml:"customPromptTemplate"`
	// The inference configuration parameters to use.
	// Default: undefined - Default inference configuration will be used.
	//
	// Experimental.
	InferenceConfig *InferenceConfiguration `field:"optional" json:"inferenceConfig" yaml:"inferenceConfig"`
	// Whether to enable or skip this step in the agent sequence.
	// Default: - The default state for each step type is as follows.
	//
	// PRE_PROCESSING – ENABLED
	// ORCHESTRATION – ENABLED
	// KNOWLEDGE_BASE_RESPONSE_GENERATION – ENABLED
	// POST_PROCESSING – DISABLED.
	//
	// Experimental.
	StepEnabled *bool `field:"optional" json:"stepEnabled" yaml:"stepEnabled"`
	// Whether to use the custom Lambda parser defined for the sequence.
	// Default: - false.
	//
	// Experimental.
	UseCustomParser *bool `field:"optional" json:"useCustomParser" yaml:"useCustomParser"`
}

Base configuration interface for all prompt step types.

Example: 

// The code below shows an example of how to instantiate this type.
// The values are placeholders you should change.
import bedrock_alpha "github.com/aws/aws-cdk-go/awsbedrockalpha"

promptStepConfigBase := &PromptStepConfigBase{
	StepType: bedrock_alpha.AgentStepType_PRE_PROCESSING,

	// the properties below are optional
	CustomPromptTemplate: jsii.String("customPromptTemplate"),
	InferenceConfig: &InferenceConfiguration{
		MaximumLength: jsii.Number(123),
		StopSequences: []*string{
			jsii.String("stopSequences"),
		},
		Temperature: jsii.Number(123),
		TopK: jsii.Number(123),
		TopP: jsii.Number(123),
	},
	StepEnabled: jsii.Boolean(false),
	UseCustomParser: jsii.Boolean(false),
}

Experimental.

type PromptTemplateConfiguration 

type PromptTemplateConfiguration interface {
}

Abstract base class for prompt template configurations.

This provides a high-level abstraction over the underlying CloudFormation template configuration properties, offering a more developer-friendly interface while maintaining full compatibility with the underlying AWS Bedrock service.

Example: 

// The code below shows an example of how to instantiate this type.
// The values are placeholders you should change.
import bedrock_alpha "github.com/aws/aws-cdk-go/awsbedrockalpha"

var chatMessage ChatMessage
var tool Tool
var toolChoice ToolChoice

promptTemplateConfiguration := bedrock_alpha.PromptTemplateConfiguration_Chat(&ChatTemplateConfigurationProps{
	Messages: []ChatMessage{
		chatMessage,
	},

	// the properties below are optional
	InputVariables: []*string{
		jsii.String("inputVariables"),
	},
	System: jsii.String("system"),
	ToolConfiguration: &ToolConfiguration{
		ToolChoice: toolChoice,
		Tools: []Tool{
			tool,
		},
	},
})

Experimental.

func PromptTemplateConfiguration_Chat 

func PromptTemplateConfiguration_Chat(props *ChatTemplateConfigurationProps) PromptTemplateConfiguration

Creates a chat template configuration. Experimental.

func PromptTemplateConfiguration_Text 

func PromptTemplateConfiguration_Text(props *TextTemplateConfigurationProps) PromptTemplateConfiguration

Creates a text template configuration. Experimental.

type PromptTemplateType 

type PromptTemplateType string

The type of prompt template. Experimental. 

const (
	// Text template for simple text-based prompts.
	// Experimental.
	PromptTemplateType_TEXT PromptTemplateType = "TEXT"
	// Chat template for conversational prompts with message history.
	// Experimental.
	PromptTemplateType_CHAT PromptTemplateType = "CHAT"
)

type PromptVariant 

type PromptVariant interface {
}

Factory class for creating prompt variants.

Provides static methods to create different types of prompt variants with proper configuration and type safety.

Example: 

cmk := kms.NewKey(this, jsii.String("cmk"), &KeyProps{
})
claudeModel := bedrock.BedrockFoundationModel_ANTHROPIC_CLAUDE_SONNET_V1_0()

variant1 := bedrock.PromptVariant_Text(&TextPromptVariantProps{
	VariantName: jsii.String("variant1"),
	Model: claudeModel,
	PromptVariables: []*string{
		jsii.String("topic"),
	},
	PromptText: jsii.String("This is my first text prompt. Please summarize our conversation on: {{topic}}."),
	InferenceConfiguration: bedrock.PromptInferenceConfiguration_Text(&PromptInferenceConfigurationProps{
		Temperature: jsii.Number(1),
		TopP: jsii.Number(0.999),
		MaxTokens: jsii.Number(2000),
	}),
})

prompt1 := bedrock.NewPrompt(this, jsii.String("prompt1"), &PromptProps{
	PromptName: jsii.String("prompt1"),
	Description: jsii.String("my first prompt"),
	DefaultVariant: variant1,
	Variants: []IPromptVariant{
		variant1,
	},
	KmsKey: cmk,
})

Experimental.

type PromptVersion 

type PromptVersion interface {
	constructs.Construct
	// The description of the prompt version.
	// Experimental.
	Description() *string
	// The tree node.
	// Experimental.
	Node() constructs.Node
	// The prompt used by this version.
	// Experimental.
	Prompt() IPrompt
	// The version of the prompt that was created.
	// Experimental.
	Version() *string
	// The Amazon Resource Name (ARN) of the prompt version.
	//
	// Example:
	//   "arn:aws:bedrock:us-east-1:123456789012:prompt/PROMPT12345:1"
	//
	// Experimental.
	VersionArn() *string
	// Returns a string representation of this construct.
	// Experimental.
	ToString() *string
	// Applies one or more mixins to this construct.
	//
	// Mixins are applied in order. The list of constructs is captured at the
	// start of the call, so constructs added by a mixin will not be visited.
	// Use multiple `with()` calls if subsequent mixins should apply to added
	// constructs.
	//
	// Returns: This construct for chaining.
	// Experimental.
	With(mixins ...constructs.IMixin) constructs.IConstruct
}

Class to create a Prompt Version with CDK.

Creates a version of the prompt. Use this to create a static snapshot of your prompt that can be deployed to production. Versions allow you to easily switch between different configurations for your prompt and update your application with the most appropriate version for your use-case.

Example: 

cmk := kms.NewKey(this, jsii.String("cmk"), &KeyProps{
})
claudeModel := bedrock.BedrockFoundationModel_ANTHROPIC_CLAUDE_SONNET_V1_0()

variant1 := bedrock.PromptVariant_Text(&TextPromptVariantProps{
	VariantName: jsii.String("variant1"),
	Model: claudeModel,
	PromptVariables: []*string{
		jsii.String("topic"),
	},
	PromptText: jsii.String("This is my first text prompt. Please summarize our conversation on: {{topic}}."),
	InferenceConfiguration: bedrock.PromptInferenceConfiguration_Text(&PromptInferenceConfigurationProps{
		Temperature: jsii.Number(1),
		TopP: jsii.Number(0.999),
		MaxTokens: jsii.Number(2000),
	}),
})

prompt1 := bedrock.NewPrompt(this, jsii.String("prompt1"), &PromptProps{
	PromptName: jsii.String("prompt1"),
	Description: jsii.String("my first prompt"),
	DefaultVariant: variant1,
	Variants: []IPromptVariant{
		variant1,
	},
	KmsKey: cmk,
})

promptVersion := bedrock.NewPromptVersion(this, jsii.String("MyPromptVersion"), &PromptVersionProps{
	Prompt: prompt1,
	Description: jsii.String("my first version"),
})
//or alternatively:
// const promptVersion = prompt1.createVersion('my first version');
versionString := promptVersion.Version

See: https://docs.aws.amazon.com/bedrock/latest/userguide/prompt-management-deploy.html

Experimental.

func NewPromptVersion 

func NewPromptVersion(scope constructs.Construct, id *string, props *PromptVersionProps) PromptVersion

**************************************************************************** CONSTRUCTOR ***************************************************************************. Experimental.

type PromptVersionProps 

type PromptVersionProps struct {
	// The prompt to use for this version.
	// Experimental.
	Prompt IPrompt `field:"required" json:"prompt" yaml:"prompt"`
	// The description of the prompt version.
	// Default: - No description provided.
	// Maximum length: 200.
	//
	// Experimental.
	Description *string `field:"optional" json:"description" yaml:"description"`
}

Properties for creating a CDK managed Bedrock Prompt Version.

Example: 

cmk := kms.NewKey(this, jsii.String("cmk"), &KeyProps{
})
claudeModel := bedrock.BedrockFoundationModel_ANTHROPIC_CLAUDE_SONNET_V1_0()

variant1 := bedrock.PromptVariant_Text(&TextPromptVariantProps{
	VariantName: jsii.String("variant1"),
	Model: claudeModel,
	PromptVariables: []*string{
		jsii.String("topic"),
	},
	PromptText: jsii.String("This is my first text prompt. Please summarize our conversation on: {{topic}}."),
	InferenceConfiguration: bedrock.PromptInferenceConfiguration_Text(&PromptInferenceConfigurationProps{
		Temperature: jsii.Number(1),
		TopP: jsii.Number(0.999),
		MaxTokens: jsii.Number(2000),
	}),
})

prompt1 := bedrock.NewPrompt(this, jsii.String("prompt1"), &PromptProps{
	PromptName: jsii.String("prompt1"),
	Description: jsii.String("my first prompt"),
	DefaultVariant: variant1,
	Variants: []IPromptVariant{
		variant1,
	},
	KmsKey: cmk,
})

promptVersion := bedrock.NewPromptVersion(this, jsii.String("MyPromptVersion"), &PromptVersionProps{
	Prompt: prompt1,
	Description: jsii.String("my first version"),
})
//or alternatively:
// const promptVersion = prompt1.createVersion('my first version');
versionString := promptVersion.Version

Experimental.

type RegexFilter 

type RegexFilter struct {
	// The action to take when a regex match is detected.
	// Experimental.
	Action GuardrailAction `field:"required" json:"action" yaml:"action"`
	// The name of the regex filter.
	// Experimental.
	Name *string `field:"required" json:"name" yaml:"name"`
	// The regular expression pattern to match.
	// Experimental.
	Pattern *string `field:"required" json:"pattern" yaml:"pattern"`
	// The description of the regex filter.
	// Default: - No description.
	//
	// Experimental.
	Description *string `field:"optional" json:"description" yaml:"description"`
	// The action to take when a regex match is detected in the input.
	// Default: GuardrailAction.BLOCK
	//
	// Experimental.
	InputAction GuardrailAction `field:"optional" json:"inputAction" yaml:"inputAction"`
	// Whether the regex filter is enabled for input.
	// Default: true.
	//
	// Experimental.
	InputEnabled *bool `field:"optional" json:"inputEnabled" yaml:"inputEnabled"`
	// The action to take when a regex match is detected in the output.
	// Default: GuardrailAction.BLOCK
	//
	// Experimental.
	OutputAction GuardrailAction `field:"optional" json:"outputAction" yaml:"outputAction"`
	// Whether the regex filter is enabled for output.
	// Default: true.
	//
	// Experimental.
	OutputEnabled *bool `field:"optional" json:"outputEnabled" yaml:"outputEnabled"`
}

A Regular expression (regex) filter for sensitive information.

Example: 

guardrail := bedrock.NewGuardrail(this, jsii.String("bedrockGuardrails"), &GuardrailProps{
	GuardrailName: jsii.String("my-BedrockGuardrails"),
})
// Add regex filter with input/output actions
guardrail.AddRegexFilter(&RegexFilter{
	Name: jsii.String("TestRegexFilter"),
	Pattern: jsii.String("test-pattern"),
	Action: bedrock.GuardrailAction_ANONYMIZE,
	// below props are optional
	Description: jsii.String("This is a test regex filter"),
	InputAction: bedrock.GuardrailAction_BLOCK,
	InputEnabled: jsii.Boolean(true),
	OutputAction: bedrock.GuardrailAction_ANONYMIZE,
	OutputEnabled: jsii.Boolean(true),
})

Experimental.

type RequireConfirmation 

type RequireConfirmation string

Enum for require confirmation state in function schemas.

Example: 

actionGroupFunction := lambda.NewFunction(this, jsii.String("ActionGroupFunction"), &FunctionProps{
	Runtime: lambda.Runtime_PYTHON_3_12(),
	Handler: jsii.String("index.handler"),
	Code: lambda.Code_FromAsset(path.join(__dirname, jsii.String("../lambda/action-group"))),
})

// Define a function schema with parameters
functionSchema := bedrock.NewFunctionSchema(&FunctionSchemaProps{
	Functions: []FunctionProps{
		&FunctionProps{
			Name: jsii.String("searchBooks"),
			Description: jsii.String("Search for books in the library catalog"),
			Parameters: map[string]FunctionParameterProps{
				"query": &FunctionParameterProps{
					"type": bedrock.ParameterType_STRING,
					"required": jsii.Boolean(true),
					"description": jsii.String("The search query string"),
				},
				"maxResults": &FunctionParameterProps{
					"type": bedrock.ParameterType_INTEGER,
					"required": jsii.Boolean(false),
					"description": jsii.String("Maximum number of results to return"),
				},
				"includeOutOfPrint": &FunctionParameterProps{
					"type": bedrock.ParameterType_BOOLEAN,
					"required": jsii.Boolean(false),
					"description": jsii.String("Whether to include out-of-print books"),
				},
			},
			RequireConfirmation: bedrock.RequireConfirmation_DISABLED,
		},
		&FunctionProps{
			Name: jsii.String("getBookDetails"),
			Description: jsii.String("Get detailed information about a specific book"),
			Parameters: map[string]FunctionParameterProps{
				"bookId": &FunctionParameterProps{
					"type": bedrock.ParameterType_STRING,
					"required": jsii.Boolean(true),
					"description": jsii.String("The unique identifier of the book"),
				},
			},
			RequireConfirmation: bedrock.RequireConfirmation_ENABLED,
		},
	},
})

// Create an action group using the function schema
actionGroup := bedrock.NewAgentActionGroup(&AgentActionGroupProps{
	Name: jsii.String("library-functions"),
	Description: jsii.String("Functions for interacting with the library catalog"),
	Executor: bedrock.ActionGroupExecutor_FromLambda(actionGroupFunction),
	FunctionSchema: functionSchema,
	Enabled: jsii.Boolean(true),
})

agent := bedrock.NewAgent(this, jsii.String("Agent"), &AgentProps{
	FoundationModel: bedrock.BedrockFoundationModel_ANTHROPIC_CLAUDE_HAIKU_V1_0(),
	Instruction: jsii.String("You are a helpful and friendly agent that answers questions about literature."),
	ActionGroups: []AgentActionGroup{
		actionGroup,
	},
})

Experimental. 

const (
	// Confirmation is enabled.
	// Experimental.
	RequireConfirmation_ENABLED RequireConfirmation = "ENABLED"
	// Confirmation is disabled.
	// Experimental.
	RequireConfirmation_DISABLED RequireConfirmation = "DISABLED"
)

type S3ApiSchema 

type S3ApiSchema interface {
	ApiSchema
	// The inline OpenAPI schema definition as a string, if using an inline schema.
	//
	// Can be in JSON or YAML format.
	// Experimental.
	InlineSchema() *string
	// The S3 location of the API schema file, if using an S3-based schema.
	//
	// Contains the bucket name and object key information.
	// Experimental.
	S3File() *awss3.Location
}

Class to define an API Schema from an S3 object.

Example: 

bucket := s3.Bucket_FromBucketName(this, jsii.String("ExistingBucket"), jsii.String("my-schema-bucket"))
s3Schema := bedrock.ApiSchema_FromS3File(bucket, jsii.String("schemas/action-group.yaml"))

actionGroupFunction := lambda.NewFunction(this, jsii.String("ActionGroupFunction"), &FunctionProps{
	Runtime: lambda.Runtime_PYTHON_3_12(),
	Handler: jsii.String("index.handler"),
	Code: lambda.Code_FromAsset(path.join(__dirname, jsii.String("../lambda/action-group"))),
})

actionGroup := bedrock.NewAgentActionGroup(&AgentActionGroupProps{
	Name: jsii.String("query-library"),
	Description: jsii.String("Use these functions to get information about the books in the library."),
	Executor: bedrock.ActionGroupExecutor_FromLambda(actionGroupFunction),
	Enabled: jsii.Boolean(true),
	ApiSchema: s3Schema,
})

agent := bedrock.NewAgent(this, jsii.String("Agent"), &AgentProps{
	FoundationModel: bedrock.BedrockFoundationModel_ANTHROPIC_CLAUDE_HAIKU_V1_0(),
	Instruction: jsii.String("You are a helpful and friendly agent that answers questions about literature."),
})

agent.AddActionGroup(actionGroup)

Experimental.

func ApiSchema_FromS3File 

func ApiSchema_FromS3File(bucket interfacesawss3.IBucketRef, objectKey *string) S3ApiSchema

Creates an API Schema from an S3 File. Experimental.

func AssetApiSchema_FromS3File 

func AssetApiSchema_FromS3File(bucket interfacesawss3.IBucketRef, objectKey *string) S3ApiSchema

Creates an API Schema from an S3 File. Experimental.

func InlineApiSchema_FromS3File 

func InlineApiSchema_FromS3File(bucket interfacesawss3.IBucketRef, objectKey *string) S3ApiSchema

Creates an API Schema from an S3 File. Experimental.

func NewS3ApiSchema 

func NewS3ApiSchema(location *awss3.Location) S3ApiSchema

Experimental.

func S3ApiSchema_FromS3File 

func S3ApiSchema_FromS3File(bucket interfacesawss3.IBucketRef, objectKey *string) S3ApiSchema

Creates an API Schema from an S3 File. Experimental.

type SessionSummaryMemoryProps 

type SessionSummaryMemoryProps struct {
	// Maximum number of recent session summaries to include (min 1).
	// Default: 20.
	//
	// Experimental.
	MaxRecentSessions *float64 `field:"optional" json:"maxRecentSessions" yaml:"maxRecentSessions"`
	// Duration for which session summaries are retained (between 1 and 365 days).
	// Default: Duration.days(30)
	//
	// Experimental.
	MemoryDuration awscdk.Duration `field:"optional" json:"memoryDuration" yaml:"memoryDuration"`
}

Properties for SessionSummaryConfiguration.

Example: 

agent := bedrock.NewAgent(this, jsii.String("MyAgent"), &AgentProps{
	AgentName: jsii.String("MyAgent"),
	Instruction: jsii.String("Your instruction here"),
	FoundationModel: bedrock.BedrockFoundationModel_AMAZON_NOVA_LITE_V1(),
	Memory: awsbedrockalpha.Memory_SessionSummary(&SessionSummaryMemoryProps{
		MaxRecentSessions: jsii.Number(10),
		 // Keep the last 10 session summaries
		MemoryDuration: awscdk.Duration_Days(jsii.Number(20)),
	}),
})

Experimental.

type TextPromptVariantProps 

type TextPromptVariantProps struct {
	// The model which is used to run the prompt.
	//
	// The model could be a foundation
	// model, a custom model, or a provisioned model.
	// Experimental.
	Model IBedrockInvokable `field:"required" json:"model" yaml:"model"`
	// The name of the prompt variant.
	// Experimental.
	VariantName *string `field:"required" json:"variantName" yaml:"variantName"`
	// The variables in the prompt template that can be filled in at runtime.
	// Default: - No variables defined.
	//
	// Experimental.
	PromptVariables *[]*string `field:"optional" json:"promptVariables" yaml:"promptVariables"`
	// The text prompt.
	//
	// Variables are used by enclosing its name with double curly braces
	// as in `{{variable_name}}`.
	// Experimental.
	PromptText *string `field:"required" json:"promptText" yaml:"promptText"`
	// Inference configuration for the Text Prompt.
	// Default: - No inference configuration provided.
	//
	// Experimental.
	InferenceConfiguration PromptInferenceConfiguration `field:"optional" json:"inferenceConfiguration" yaml:"inferenceConfiguration"`
}

Properties for creating a text prompt variant.

Example: 

cmk := kms.NewKey(this, jsii.String("cmk"), &KeyProps{
})
claudeModel := bedrock.BedrockFoundationModel_ANTHROPIC_CLAUDE_SONNET_V1_0()

variant1 := bedrock.PromptVariant_Text(&TextPromptVariantProps{
	VariantName: jsii.String("variant1"),
	Model: claudeModel,
	PromptVariables: []*string{
		jsii.String("topic"),
	},
	PromptText: jsii.String("This is my first text prompt. Please summarize our conversation on: {{topic}}."),
	InferenceConfiguration: bedrock.PromptInferenceConfiguration_Text(&PromptInferenceConfigurationProps{
		Temperature: jsii.Number(1),
		TopP: jsii.Number(0.999),
		MaxTokens: jsii.Number(2000),
	}),
})

prompt1 := bedrock.NewPrompt(this, jsii.String("prompt1"), &PromptProps{
	PromptName: jsii.String("prompt1"),
	Description: jsii.String("my first prompt"),
	DefaultVariant: variant1,
	Variants: []IPromptVariant{
		variant1,
	},
	KmsKey: cmk,
})

Experimental.

type TextTemplateConfigurationProps 

type TextTemplateConfigurationProps struct {
	// The text content of the template.
	// Experimental.
	Text *string `field:"required" json:"text" yaml:"text"`
	// The input variables for the template.
	// Default: - No input variables.
	//
	// Experimental.
	InputVariables *[]*string `field:"optional" json:"inputVariables" yaml:"inputVariables"`
}

Properties for creating a text template configuration.

Example: 

// The code below shows an example of how to instantiate this type.
// The values are placeholders you should change.
import bedrock_alpha "github.com/aws/aws-cdk-go/awsbedrockalpha"

textTemplateConfigurationProps := &TextTemplateConfigurationProps{
	Text: jsii.String("text"),

	// the properties below are optional
	InputVariables: []*string{
		jsii.String("inputVariables"),
	},
}

Experimental.

type TierConfig 

type TierConfig string

**************************************************************************** TIER CONFIG ***************************************************************************.

Example: 

guardrail := bedrock.NewGuardrail(this, jsii.String("bedrockGuardrails"), &GuardrailProps{
	GuardrailName: jsii.String("my-BedrockGuardrails"),
	// Configure tier for topic filters (optional)
	TopicsTierConfig: bedrock.TierConfig_STANDARD,
})

// Use a predefined topic
guardrail.AddDeniedTopicFilter(bedrock.Topic_FINANCIAL_ADVICE())

// Create a custom topic with input/output actions
guardrail.AddDeniedTopicFilter(bedrock.Topic_Custom(&CustomTopicProps{
	Name: jsii.String("Legal_Advice"),
	Definition: jsii.String("Offering guidance or suggestions on legal matters, legal actions, interpretation of laws, or legal rights and responsibilities."),
	Examples: []*string{
		jsii.String("Can I sue someone for this?"),
		jsii.String("What are my legal rights in this situation?"),
		jsii.String("Is this action against the law?"),
		jsii.String("What should I do to file a legal complaint?"),
		jsii.String("Can you explain this law to me?"),
	},
	// props below are optional
	InputAction: bedrock.GuardrailAction_BLOCK,
	InputEnabled: jsii.Boolean(true),
	OutputAction: bedrock.GuardrailAction_NONE,
	OutputEnabled: jsii.Boolean(true),
}))

Experimental. 

const (
	// Provides established guardrails functionality supporting English, French, and Spanish languages.
	// Experimental.
	TierConfig_CLASSIC TierConfig = "CLASSIC"
	// Provides a more robust solution than the CLASSIC tier and has more comprehensive language support.
	//
	// This tier requires that your guardrail use cross-Region inference.
	// Experimental.
	TierConfig_STANDARD TierConfig = "STANDARD"
)

type Tool 

type Tool interface {
}

Abstract base class for tools that can be used by the model.

Example: 

cmk := kms.NewKey(this, jsii.String("cmk"), &KeyProps{
})

variantChat := bedrock.PromptVariant_Chat(&ChatPromptVariantProps{
	VariantName: jsii.String("variant1"),
	Model: bedrock.BedrockFoundationModel_ANTHROPIC_CLAUDE_3_5_SONNET_V1_0(),
	Messages: []ChatMessage{
		bedrock.ChatMessage_User(jsii.String("From now on, you speak Japanese!")),
		bedrock.ChatMessage_Assistant(jsii.String("Konnichiwa!")),
		bedrock.ChatMessage_*User(jsii.String("From now on, you speak {{language}}!")),
	},
	System: jsii.String("You are a helpful assistant that only speaks the language you`re told."),
	PromptVariables: []*string{
		jsii.String("language"),
	},
	ToolConfiguration: &ToolConfiguration{
		ToolChoice: bedrock.ToolChoice_AUTO(),
		Tools: []Tool{
			bedrock.Tool_Function(&FunctionToolProps{
				Name: jsii.String("top_song"),
				Description: jsii.String("Get the most popular song played on a radio station."),
				InputSchema: map[string]interface{}{
					"type": jsii.String("object"),
					"properties": map[string]map[string]*string{
						"sign": map[string]*string{
							"type": jsii.String("string"),
							"description": jsii.String("The call sign for the radio station for which you want the most popular song. Example calls signs are WZPZ and WKR."),
						},
					},
					"required": []*string{
						jsii.String("sign"),
					},
				},
			}),
		},
	},
})

bedrock.NewPrompt(this, jsii.String("prompt1"), &PromptProps{
	PromptName: jsii.String("prompt-chat"),
	Description: jsii.String("my first chat prompt"),
	DefaultVariant: variantChat,
	Variants: []IPromptVariant{
		variantChat,
	},
	KmsKey: cmk,
})

Experimental.

func Tool_Function 

func Tool_Function(props *FunctionToolProps) Tool

Creates a function tool. Experimental.

type ToolChoice 

type ToolChoice interface {
	// Configuration for ANY tool choice.
	// Experimental.
	Any() interface{}
	// Configuration for AUTO tool choice.
	// Experimental.
	Auto() interface{}
	// The specific tool name if using specific tool choice.
	// Experimental.
	Tool() *string
}

Defines how the model should choose which tool to use.

Example: 

cmk := kms.NewKey(this, jsii.String("cmk"), &KeyProps{
})

variantChat := bedrock.PromptVariant_Chat(&ChatPromptVariantProps{
	VariantName: jsii.String("variant1"),
	Model: bedrock.BedrockFoundationModel_ANTHROPIC_CLAUDE_3_5_SONNET_V1_0(),
	Messages: []ChatMessage{
		bedrock.ChatMessage_User(jsii.String("From now on, you speak Japanese!")),
		bedrock.ChatMessage_Assistant(jsii.String("Konnichiwa!")),
		bedrock.ChatMessage_*User(jsii.String("From now on, you speak {{language}}!")),
	},
	System: jsii.String("You are a helpful assistant that only speaks the language you`re told."),
	PromptVariables: []*string{
		jsii.String("language"),
	},
	ToolConfiguration: &ToolConfiguration{
		ToolChoice: bedrock.ToolChoice_AUTO(),
		Tools: []Tool{
			bedrock.Tool_Function(&FunctionToolProps{
				Name: jsii.String("top_song"),
				Description: jsii.String("Get the most popular song played on a radio station."),
				InputSchema: map[string]interface{}{
					"type": jsii.String("object"),
					"properties": map[string]map[string]*string{
						"sign": map[string]*string{
							"type": jsii.String("string"),
							"description": jsii.String("The call sign for the radio station for which you want the most popular song. Example calls signs are WZPZ and WKR."),
						},
					},
					"required": []*string{
						jsii.String("sign"),
					},
				},
			}),
		},
	},
})

bedrock.NewPrompt(this, jsii.String("prompt1"), &PromptProps{
	PromptName: jsii.String("prompt-chat"),
	Description: jsii.String("my first chat prompt"),
	DefaultVariant: variantChat,
	Variants: []IPromptVariant{
		variantChat,
	},
	KmsKey: cmk,
})

Experimental.

func NewToolChoice 

func NewToolChoice(any interface{}, auto interface{}, tool *string) ToolChoice

Experimental.

func ToolChoice_ANY 

func ToolChoice_ANY() ToolChoice

func ToolChoice_AUTO 

func ToolChoice_AUTO() ToolChoice

func ToolChoice_SpecificTool 

func ToolChoice_SpecificTool(toolName *string) ToolChoice

The Model must request the specified tool.

Only supported by some models like Anthropic Claude 3 models.

Returns: A ToolChoice instance configured for the specific tool. Experimental.

type ToolConfiguration 

type ToolConfiguration struct {
	// How the model should choose which tool to use.
	// Experimental.
	ToolChoice ToolChoice `field:"required" json:"toolChoice" yaml:"toolChoice"`
	// The tools available to the model.
	// Experimental.
	Tools *[]Tool `field:"required" json:"tools" yaml:"tools"`
}

Configuration for tools available to the model.

Example: 

cmk := kms.NewKey(this, jsii.String("cmk"), &KeyProps{
})

variantChat := bedrock.PromptVariant_Chat(&ChatPromptVariantProps{
	VariantName: jsii.String("variant1"),
	Model: bedrock.BedrockFoundationModel_ANTHROPIC_CLAUDE_3_5_SONNET_V1_0(),
	Messages: []ChatMessage{
		bedrock.ChatMessage_User(jsii.String("From now on, you speak Japanese!")),
		bedrock.ChatMessage_Assistant(jsii.String("Konnichiwa!")),
		bedrock.ChatMessage_*User(jsii.String("From now on, you speak {{language}}!")),
	},
	System: jsii.String("You are a helpful assistant that only speaks the language you`re told."),
	PromptVariables: []*string{
		jsii.String("language"),
	},
	ToolConfiguration: &ToolConfiguration{
		ToolChoice: bedrock.ToolChoice_AUTO(),
		Tools: []Tool{
			bedrock.Tool_Function(&FunctionToolProps{
				Name: jsii.String("top_song"),
				Description: jsii.String("Get the most popular song played on a radio station."),
				InputSchema: map[string]interface{}{
					"type": jsii.String("object"),
					"properties": map[string]map[string]*string{
						"sign": map[string]*string{
							"type": jsii.String("string"),
							"description": jsii.String("The call sign for the radio station for which you want the most popular song. Example calls signs are WZPZ and WKR."),
						},
					},
					"required": []*string{
						jsii.String("sign"),
					},
				},
			}),
		},
	},
})

bedrock.NewPrompt(this, jsii.String("prompt1"), &PromptProps{
	PromptName: jsii.String("prompt-chat"),
	Description: jsii.String("my first chat prompt"),
	DefaultVariant: variantChat,
	Variants: []IPromptVariant{
		variantChat,
	},
	KmsKey: cmk,
})

Experimental.

type Topic 

type Topic interface {
	// Definition of the topic.
	// Experimental.
	Definition() *string
	// Representative phrases that refer to the topic.
	// Experimental.
	Examples() *[]*string
	// The action to take when a topic is detected in the input.
	// Default: GuardrailAction.BLOCK
	//
	// Experimental.
	InputAction() GuardrailAction
	// Whether the topic filter is enabled for input.
	// Default: true.
	//
	// Experimental.
	InputEnabled() *bool
	// The name of the topic to deny.
	// Experimental.
	Name() *string
	// The action to take when a topic is detected in the output.
	// Default: GuardrailAction.BLOCK
	//
	// Experimental.
	OutputAction() GuardrailAction
	// Whether the topic filter is enabled for output.
	// Default: true.
	//
	// Experimental.
	OutputEnabled() *bool
}

Represents predefined topics that can be used to filter content.

Example: 

guardrail := bedrock.NewGuardrail(this, jsii.String("bedrockGuardrails"), &GuardrailProps{
	GuardrailName: jsii.String("my-BedrockGuardrails"),
	// Configure tier for topic filters (optional)
	TopicsTierConfig: bedrock.TierConfig_STANDARD,
})

// Use a predefined topic
guardrail.AddDeniedTopicFilter(bedrock.Topic_FINANCIAL_ADVICE())

// Create a custom topic with input/output actions
guardrail.AddDeniedTopicFilter(bedrock.Topic_Custom(&CustomTopicProps{
	Name: jsii.String("Legal_Advice"),
	Definition: jsii.String("Offering guidance or suggestions on legal matters, legal actions, interpretation of laws, or legal rights and responsibilities."),
	Examples: []*string{
		jsii.String("Can I sue someone for this?"),
		jsii.String("What are my legal rights in this situation?"),
		jsii.String("Is this action against the law?"),
		jsii.String("What should I do to file a legal complaint?"),
		jsii.String("Can you explain this law to me?"),
	},
	// props below are optional
	InputAction: bedrock.GuardrailAction_BLOCK,
	InputEnabled: jsii.Boolean(true),
	OutputAction: bedrock.GuardrailAction_NONE,
	OutputEnabled: jsii.Boolean(true),
}))

Experimental.

func NewTopic 

func NewTopic(props *CustomTopicProps) Topic

Experimental.

func Topic_Custom 

func Topic_Custom(props *CustomTopicProps) Topic

Create a custom topic filter. Experimental.

func Topic_FINANCIAL_ADVICE 

func Topic_FINANCIAL_ADVICE() Topic

func Topic_INAPPROPRIATE_CONTENT 

func Topic_INAPPROPRIATE_CONTENT() Topic
func Topic_LEGAL_ADVICE() Topic

func Topic_MEDICAL_ADVICE 

func Topic_MEDICAL_ADVICE() Topic

func Topic_POLITICAL_ADVICE 

func Topic_POLITICAL_ADVICE() Topic

type UKSpecificPIIType 

type UKSpecificPIIType interface {
	PIIType
	// The string value of the PII type.
	// Experimental.
	Value() *string
	// Returns the string representation of the PII type.
	//
	// Returns: The string value of the PII type.
	// Experimental.
	ToString() *string
}

Types of PII specific to the United Kingdom (UK).

Example: 

// The code below shows an example of how to instantiate this type.
// The values are placeholders you should change.
import bedrock_alpha "github.com/aws/aws-cdk-go/awsbedrockalpha"

uKSpecificPIIType := bedrock_alpha.UKSpecificPIIType_UK_NATIONAL_HEALTH_SERVICE_NUMBER()

Experimental.

func UKSpecificPIIType_UK_NATIONAL_HEALTH_SERVICE_NUMBER 

func UKSpecificPIIType_UK_NATIONAL_HEALTH_SERVICE_NUMBER() UKSpecificPIIType

func UKSpecificPIIType_UK_NATIONAL_INSURANCE_NUMBER 

func UKSpecificPIIType_UK_NATIONAL_INSURANCE_NUMBER() UKSpecificPIIType

func UKSpecificPIIType_UK_UNIQUE_TAXPAYER_REFERENCE_NUMBER 

func UKSpecificPIIType_UK_UNIQUE_TAXPAYER_REFERENCE_NUMBER() UKSpecificPIIType

type USASpecificPIIType 

type USASpecificPIIType interface {
	PIIType
	// The string value of the PII type.
	// Experimental.
	Value() *string
	// Returns the string representation of the PII type.
	//
	// Returns: The string value of the PII type.
	// Experimental.
	ToString() *string
}

Types of PII specific to the USA.

Example: 

guardrail := bedrock.NewGuardrail(this, jsii.String("bedrockGuardrails"), &GuardrailProps{
	GuardrailName: jsii.String("my-BedrockGuardrails"),
})

// Add PII filter for addresses with input/output actions
guardrail.AddPIIFilter(&PIIFilter{
	Type: bedrock.GeneralPIIType_ADDRESS(),
	Action: bedrock.GuardrailAction_BLOCK,
	// below props are optional
	InputAction: bedrock.GuardrailAction_BLOCK,
	InputEnabled: jsii.Boolean(true),
	OutputAction: bedrock.GuardrailAction_ANONYMIZE,
	OutputEnabled: jsii.Boolean(true),
})

// Add PII filter for credit card numbers with input/output actions
guardrail.AddPIIFilter(&PIIFilter{
	Type: bedrock.FinancePIIType_CREDIT_DEBIT_CARD_NUMBER(),
	Action: bedrock.GuardrailAction_BLOCK,
	// below props are optional
	InputAction: bedrock.GuardrailAction_BLOCK,
	InputEnabled: jsii.Boolean(true),
	OutputAction: bedrock.GuardrailAction_ANONYMIZE,
	OutputEnabled: jsii.Boolean(true),
})

// Add PII filter for email addresses
guardrail.AddPIIFilter(&PIIFilter{
	Type: bedrock.GeneralPIIType_EMAIL(),
	Action: bedrock.GuardrailAction_ANONYMIZE,
})

// Add PII filter for US Social Security Numbers
guardrail.AddPIIFilter(&PIIFilter{
	Type: bedrock.USASpecificPIIType_US_SOCIAL_SECURITY_NUMBER(),
	Action: bedrock.GuardrailAction_BLOCK,
})

// Add PII filter for IP addresses
guardrail.AddPIIFilter(&PIIFilter{
	Type: bedrock.InformationTechnologyPIIType_IP_ADDRESS(),
	Action: bedrock.GuardrailAction_ANONYMIZE,
})

Experimental.

func USASpecificPIIType_US_BANK_ACCOUNT_NUMBER 

func USASpecificPIIType_US_BANK_ACCOUNT_NUMBER() USASpecificPIIType

func USASpecificPIIType_US_BANK_ROUTING_NUMBER 

func USASpecificPIIType_US_BANK_ROUTING_NUMBER() USASpecificPIIType

func USASpecificPIIType_US_INDIVIDUAL_TAX_IDENTIFICATION_NUMBER 

func USASpecificPIIType_US_INDIVIDUAL_TAX_IDENTIFICATION_NUMBER() USASpecificPIIType

func USASpecificPIIType_US_PASSPORT_NUMBER 

func USASpecificPIIType_US_PASSPORT_NUMBER() USASpecificPIIType

func USASpecificPIIType_US_SOCIAL_SECURITY_NUMBER 

func USASpecificPIIType_US_SOCIAL_SECURITY_NUMBER() USASpecificPIIType

type VectorType 

type VectorType string

The data type for the vectors when using a model to convert text into vector embeddings.

The model must support the specified data type for vector embeddings. Floating-point (float32) is the default data type, and is supported by most models for vector embeddings. See Supported embeddings models for information on the available models and their vector data types. Experimental. 

const (
	// `FLOATING_POINT` convert the data to floating-point (float32) vector embeddings (more precise, but more costly).
	// Experimental.
	VectorType_FLOATING_POINT VectorType = "FLOATING_POINT"
	// `BINARY` convert the data to binary vector embeddings (less precise, but less costly).
	// Experimental.
	VectorType_BINARY VectorType = "BINARY"
)

type WordFilter 

type WordFilter struct {
	// The text to filter.
	// Experimental.
	Text *string `field:"required" json:"text" yaml:"text"`
	// The action to take when a word is detected in the input.
	// Default: GuardrailAction.BLOCK
	//
	// Experimental.
	InputAction GuardrailAction `field:"optional" json:"inputAction" yaml:"inputAction"`
	// Whether the word filter is enabled for input.
	// Default: true.
	//
	// Experimental.
	InputEnabled *bool `field:"optional" json:"inputEnabled" yaml:"inputEnabled"`
	// The action to take when a word is detected in the output.
	// Default: GuardrailAction.BLOCK
	//
	// Experimental.
	OutputAction GuardrailAction `field:"optional" json:"outputAction" yaml:"outputAction"`
	// Whether the word filter is enabled for output.
	// Default: true.
	//
	// Experimental.
	OutputEnabled *bool `field:"optional" json:"outputEnabled" yaml:"outputEnabled"`
}

Interface to define a Word Filter.

Example: 

guardrail := bedrock.NewGuardrail(this, jsii.String("bedrockGuardrails"), &GuardrailProps{
	GuardrailName: jsii.String("my-BedrockGuardrails"),
})

// Add managed word list with input/output actions
guardrail.AddManagedWordListFilter(&ManagedWordFilter{
	Type: bedrock.ManagedWordFilterType_PROFANITY,
	InputAction: bedrock.GuardrailAction_BLOCK,
	InputEnabled: jsii.Boolean(true),
	OutputAction: bedrock.GuardrailAction_NONE,
	OutputEnabled: jsii.Boolean(true),
})

// Add individual words
guardrail.AddWordFilter(&WordFilter{
	Text: jsii.String("drugs"),
})
guardrail.AddWordFilter(&WordFilter{
	Text: jsii.String("competitor"),
})

// Add words from a file
guardrail.AddWordFilterFromFile(jsii.String("./scripts/wordsPolicy.csv"))

Experimental.

Source Files

View all Source files

Directories

Path Synopsis
Package jsii contains the functionaility needed for jsii packages to initialize their dependencies and themselves.
 Package jsii contains the functionaility needed for jsii packages to initialize their dependencies and themselves.

Jump to

Keyboard shortcuts

? : This menu
/ : Search site
f or F : Jump to
y or Y : Canonical URL
go.dev uses cookies from Google to deliver and enhance the quality of its services and to analyze traffic. Learn more.