@aws-cdk/aws-bedrock-agentcore-alpha 2.234.0-alpha.0 → 2.235.0-alpha.0

package/.jsii

@@ -8,8 +8,8 @@
     "url": "https://aws.amazon.com"
   },
   "dependencies": {
-    "@aws-cdk/aws-bedrock-alpha": "2.234.0-alpha.0",
-    "aws-cdk-lib": "^2.234.0",
+    "@aws-cdk/aws-bedrock-alpha": "2.235.0-alpha.0",
+    "aws-cdk-lib": "^2.235.0",
     "constructs": "^10.0.0"
   },
   "dependencyClosure": {
@@ -8524,7 +8524,7 @@
   },
   "name": "@aws-cdk/aws-bedrock-agentcore-alpha",
   "readme": {
-    "markdown": "# Amazon Bedrock AgentCore Construct Library\n\n<!--BEGIN STABILITY BANNER-->\n\n---\n\n![cdk-constructs: Experimental](https://img.shields.io/badge/cdk--constructs-experimental-important.svg?style=for-the-badge)\n\n> The APIs of higher level constructs in this module are experimental and under active development.\n> They are subject to non-backward compatible changes or removal in any future version. These are\n> not subject to the [Semantic Versioning](https://semver.org/) model and breaking changes will be\n> announced in the release notes. This means that while you may use them, you may need to update\n> your source code when upgrading to a newer version of this package.\n\n---\n\n<!--END STABILITY BANNER-->\n\n| **Language**                                                                                   | **Package**                             |\n| :--------------------------------------------------------------------------------------------- | --------------------------------------- |\n| ![Typescript Logo](https://docs.aws.amazon.com/cdk/api/latest/img/typescript32.png) TypeScript | `@aws-cdk/aws-bedrock-agentcore-alpha` |\n\n[Amazon Bedrock AgentCore](https://aws.amazon.com/bedrock/agentcore/) enables you to deploy and operate highly capable AI agents securely, at scale. It offers infrastructure purpose-built for dynamic agent workloads, powerful tools to enhance agents, and essential controls for real-world deployment. AgentCore services can be used together or independently and work with any framework including CrewAI, LangGraph, LlamaIndex, and Strands Agents, as well as any foundation model in or outside of Amazon Bedrock, giving you ultimate flexibility. AgentCore eliminates the undifferentiated heavy lifting of building specialized agent infrastructure, so you can accelerate agents to production.\n\nThis construct library facilitates the deployment of Bedrock AgentCore primitives, enabling you to create sophisticated AI applications that can interact with your systems and data sources.\n\n> **Note:** Users need to ensure their CDK deployment role has the `iam:CreateServiceLinkedRole` permission for AgentCore service-linked roles.\n\n## Table of contents\n\n- [AgentCore Runtime](#agentcore-runtime)\n  - [Runtime Versioning](#runtime-versioning)\n  - [Runtime Endpoints](#runtime-endpoints)\n  - [AgentCore Runtime Properties](#agentcore-runtime-properties)\n  - [Runtime Endpoint Properties](#runtime-endpoint-properties)\n  - [Creating a Runtime](#creating-a-runtime)\n    - [Option 1: Use an existing image in ECR](#option-1-use-an-existing-image-in-ecr)\n    - [Managing Endpoints and Versions](#managing-endpoints-and-versions)\n    - [Option 2: Use a local asset](#option-2-use-a-local-asset)\n- [Browser Custom tool](#browser)\n  - [Browser properties](#browser-properties)\n  - [Browser Network modes](#browser-network-modes)\n  - [Basic Browser Creation](#basic-browser-creation)\n  - [Browser IAM permissions](#browser-iam-permissions)\n- [Code Interpreter Custom tool](#code-interpreter)\n  - [Code Interpreter properties](#code-interpreter-properties)\n  - [Code Interpreter Network Modes](#code-interpreter-network-modes)\n  - [Basic Code Interpreter Creation](#basic-code-interpreter-creation)\n  - [Code Interpreter IAM permissions](#code-interpreter-iam-permissions)\n- [Gateway](#gateway)\n  - [Gateway Properties](#gateway-properties)\n  - [Basic Gateway Creation](#basic-gateway-creation)\n  - [Protocol configuration](#protocol-configuration)\n  - [Inbound authorization](#inbound-authorization)\n  - [Gateway with KMS Encryption](#gateway-with-kms-encryption)\n  - [Gateway with Custom Execution Role](#gateway-with-custom-execution-role)\n  - [Gateway IAM Permissions](#gateway-iam-permissions)\n- [Gateway Target](#gateway-target)\n  - [Gateway Target Properties](#gateway-target-properties)\n  - [Targets types](#targets-types)\n  - [Outbound auth](#outbound-auth)\n  - [Api schema](#api-schema-for-openapi-and-smithy-target)\n  - [Basic Gateway Target Creation](#basic-gateway-target-creation)\n    - [Using addTarget methods (Recommended)](#using-addtarget-methods-recommended)\n    - [Using static factory methods](#using-static-factory-methods)\n  - [Lambda Target with Tool Schema](#tools-schema-for-lambda-target)\n  - [Smithy Model Target with OAuth](#api-schema-for-openapi-and-smithy-target)\n  - [Gateway Target IAM Permissions](#gateway-target-iam-permissions)\n- [Memory](#memory)\n  - [Memory properties](#memory-properties)\n  - [Basic Memory Creation](#basic-memory-creation)\n  - [LTM Memory Extraction Stategies](#ltm-memory-extraction-stategies)\n  - [Memory Strategy Methods](#memory-strategy-methods)\n- [Amazon Bedrock AgentCore Construct Library](#amazon-bedrock-agentcore-construct-library)\n  - [Table of contents](#table-of-contents)\n  - [AgentCore Runtime](#agentcore-runtime)\n    - [Runtime Endpoints](#runtime-endpoints)\n    - [AgentCore Runtime Properties](#agentcore-runtime-properties)\n    - [Runtime Endpoint Properties](#runtime-endpoint-properties)\n    - [Creating a Runtime](#creating-a-runtime)\n      - [Option 1: Use an existing image in ECR](#option-1-use-an-existing-image-in-ecr)\n      - [Option 2: Use a local asset](#option-2-use-a-local-asset)\n      - [Option 3: Use direct code deployment](#option-3-use-direct-code-deployment)\n    - [Granting Permissions to Invoke Bedrock Models or Inference Profiles](#granting-permissions-to-invoke-bedrock-models-or-inference-profiles)\n    - [Runtime Versioning](#runtime-versioning)\n      - [Managing Endpoints and Versions](#managing-endpoints-and-versions)\n        - [Step 1: Initial Deployment](#step-1-initial-deployment)\n        - [Step 2: Creating Custom Endpoints](#step-2-creating-custom-endpoints)\n        - [Step 3: Runtime Update Deployment](#step-3-runtime-update-deployment)\n        - [Step 4: Testing with Staging Endpoints](#step-4-testing-with-staging-endpoints)\n        - [Step 5: Promoting to Production](#step-5-promoting-to-production)\n    - [Creating Standalone Runtime Endpoints](#creating-standalone-runtime-endpoints)\n      - [Example: Creating an endpoint for an existing runtime](#example-creating-an-endpoint-for-an-existing-runtime)\n    - [Runtime Authentication Configuration](#runtime-authentication-configuration)\n      - [IAM Authentication (Default)](#iam-authentication-default)\n      - [Cognito Authentication](#cognito-authentication)\n      - [JWT Authentication](#jwt-authentication)\n      - [OAuth Authentication](#oauth-authentication)\n      - [Using a Custom IAM Role](#using-a-custom-iam-role)\n    - [Runtime Network Configuration](#runtime-network-configuration)\n      - [Public Network Mode (Default)](#public-network-mode-default)\n      - [VPC Network Mode](#vpc-network-mode)\n      - [Managing Security Groups with VPC Configuration](#managing-security-groups-with-vpc-configuration)\n  - [Browser](#browser)\n    - [Browser Network modes](#browser-network-modes)\n    - [Browser Properties](#browser-properties)\n    - [Basic Browser Creation](#basic-browser-creation)\n    - [Browser with Tags](#browser-with-tags)\n    - [Browser with VPC](#browser-with-vpc)\n    - [Browser with Recording Configuration](#browser-with-recording-configuration)\n    - [Browser with Custom Execution Role](#browser-with-custom-execution-role)\n    - [Browser with S3 Recording and Permissions](#browser-with-s3-recording-and-permissions)\n    - [Browser IAM Permissions](#browser-iam-permissions)\n  - [Code Interpreter](#code-interpreter)\n    - [Code Interpreter Network Modes](#code-interpreter-network-modes)\n    - [Code Interpreter Properties](#code-interpreter-properties)\n    - [Basic Code Interpreter Creation](#basic-code-interpreter-creation)\n    - [Code Interpreter with VPC](#code-interpreter-with-vpc)\n    - [Code Interpreter with Sandbox Network Mode](#code-interpreter-with-sandbox-network-mode)\n    - [Code Interpreter with Custom Execution Role](#code-interpreter-with-custom-execution-role)\n    - [Code Interpreter IAM Permissions](#code-interpreter-iam-permissions)\n    - [Code interpreter with tags](#code-interpreter-with-tags)\n  - [Memory](#memory)\n    - [Memory Properties](#memory-properties)\n    - [Basic Memory Creation](#basic-memory-creation)\n    - [LTM Memory Extraction Stategies](#ltm-memory-extraction-stategies)\n    - [Memory with Built-in Strategies](#memory-with-built-in-strategies)\n    - [Memory with custom Strategies](#memory-with-custom-strategies)\n      - [Memory with Custom Execution Role](#memory-with-custom-execution-role)\n    - [Memory with self-managed Strategies](#memory-with-self-managed-strategies)\n    - [Memory Strategy Methods](#memory-strategy-methods)\n\n## AgentCore Runtime\n\nThe AgentCore Runtime construct enables you to deploy containerized agents on Amazon Bedrock AgentCore.\nThis L2 construct simplifies runtime creation just pass your ECR repository name\nand the construct handles all the configuration with sensible defaults.\n\n### Runtime Endpoints\n\nEndpoints provide a stable way to invoke specific versions of your agent runtime, enabling controlled deployments across different environments.\nWhen you create an agent runtime, Amazon Bedrock AgentCore automatically creates a \"DEFAULT\" endpoint which always points to the latest version\nof runtime.\n\nYou can create additional endpoints in two ways:\n\n1. **Using Runtime.addEndpoint()** - Convenient method when creating endpoints alongside the runtime.\n2. **Using RuntimeEndpoint** - Flexible approach for existing runtimes.\n\nFor example, you might keep a \"production\" endpoint on a stable version while testing newer versions\nthrough a \"staging\" endpoint. This separation allows you to test changes thoroughly before promoting them\nto production by simply updating the endpoint to point to the newer version.\n\n### AgentCore Runtime Properties\n\n| Name | Type | Required | Description |\n|------|------|----------|-------------|\n| `runtimeName` | `string` | Yes | The name of the agent runtime. Valid characters are a-z, A-Z, 0-9, _ (underscore). Must start with a letter and can be up to 48 characters long |\n| `agentRuntimeArtifact` | `AgentRuntimeArtifact` | Yes | The artifact configuration for the agent runtime containing the container configuration with ECR URI |\n| `executionRole` | `iam.IRole` | No | The IAM role that provides permissions for the agent runtime. If not provided, a role will be created automatically |\n| `networkConfiguration` | `NetworkConfiguration` | No | Network configuration for the agent runtime. Defaults to `RuntimeNetworkConfiguration.usingPublicNetwork()` |\n| `description` | `string` | No | Optional description for the agent runtime |\n| `protocolConfiguration` | `ProtocolType` | No | Protocol configuration for the agent runtime. Defaults to `ProtocolType.HTTP` |\n| `authorizerConfiguration` | `RuntimeAuthorizerConfiguration` | No | Authorizer configuration for the agent runtime. Use `RuntimeAuthorizerConfiguration` static methods to create configurations for IAM, Cognito, JWT, or OAuth authentication |\n| `environmentVariables` | `{ [key: string]: string }` | No | Environment variables for the agent runtime. Maximum 50 environment variables |\n| `tags` | `{ [key: string]: string }` | No | Tags for the agent runtime. A list of key:value pairs of tags to apply to this Runtime resource |\n| `lifecycleConfiguration` | LifecycleConfiguration | No | The life cycle configuration for the AgentCore Runtime. Defaults to 900 seconds (15 minutes) for idle, 28800 seconds (8 hours) for max life time |\n| `requestHeaderConfiguration` | RequestHeaderConfiguration | No | Configuration for HTTP request headers that will be passed through to the runtime. Defaults to no configuration |\n\n### Runtime Endpoint Properties\n\n| Name | Type | Required | Description |\n|------|------|----------|-------------|\n| `endpointName` | `string` | Yes | The name of the runtime endpoint. Valid characters are a-z, A-Z, 0-9, _ (underscore). Must start with a letter and can be up to 48 characters long |\n| `agentRuntimeId` | `string` | Yes | The Agent Runtime ID for this endpoint |\n| `agentRuntimeVersion` | `string` | Yes | The Agent Runtime version for this endpoint. Must be between 1 and 5 characters long.|\n| `description` | `string` | No | Optional description for the runtime endpoint |\n| `tags` | `{ [key: string]: string }` | No | Tags for the runtime endpoint |\n\n### Creating a Runtime\n\n#### Option 1: Use an existing image in ECR\n\nReference an image available within ECR.\n\n```typescript fixture=default\nconst repository = new ecr.Repository(this, \"TestRepository\", {\n  repositoryName: \"test-agent-runtime\",\n});\n\n// The runtime by default create ECR permission only for the repository available in the account the stack is being deployed\nconst agentRuntimeArtifact = agentcore.AgentRuntimeArtifact.fromEcrRepository(repository, \"v1.0.0\");\n\n// Create runtime using the built image\nconst runtime = new agentcore.Runtime(this, \"MyAgentRuntime\", {\n  runtimeName: \"myAgent\",\n  agentRuntimeArtifact: agentRuntimeArtifact\n});\n```\n\n#### Option 2: Use a local asset\n\nReference a local directory containing a Dockerfile.\nImages are built from a local Docker context directory (with a Dockerfile), uploaded to Amazon Elastic Container Registry (ECR)\nby the CDK toolkit,and can be naturally referenced in your CDK app.\n\n```typescript fixture=default\nconst agentRuntimeArtifact = agentcore.AgentRuntimeArtifact.fromAsset(\n  path.join(__dirname, \"path to agent dockerfile directory\")\n);\n\nconst runtime = new agentcore.Runtime(this, \"MyAgentRuntime\", {\n  runtimeName: \"myAgent\",\n  agentRuntimeArtifact: agentRuntimeArtifact,\n});\n```\n\n#### Option 3: Use direct code deployment\n\nWith the container deployment method, developers create a Dockerfile, build ARM-compatible containers, manage ECR repositories, and upload containers for code changes. This works well where container DevOps pipelines have already been established to automate deployments.\n\nHowever, customers looking for fully managed deployments can benefit from direct code deployment, which can significantly improve developer time and productivity. Direct code deployment provides a secure and scalable path forward for rapid prototyping agent capabilities to deploying production workloads at scale.\n\nWith direct code deployment, developers create a zip archive of code and dependencies, upload to Amazon S3, and configure the bucket in the agent configuration. A ZIP archive containing Linux arm64 dependencies needs to be uploaded to S3 as a pre-requisite to Create Agent Runtime.\n\nFor more information, please refer to the [documentation](https://docs.aws.amazon.com/bedrock-agentcore/latest/devguide/runtime-get-started-code-deploy.html).\n\n```typescript fixture=default\n// S3 bucket containing the agent core\nconst codeBucket = new s3.Bucket(this, \"AgentCode\", {\n  bucketName: \"my-code-bucket\",\n  removalPolicy: RemovalPolicy.DESTROY, // For demo purposes\n});\n\n// the bucket above needs to contain the agent code\n\nconst agentRuntimeArtifact = agentcore.AgentRuntimeArtifact.fromS3(\n  {\n    bucketName: codeBucket.bucketName,\n    objectKey: 'deployment_package.zip',\n  },\n  agentcore.AgentCoreRuntime.PYTHON_3_12,\n  ['opentelemetry-instrument', 'main.py']\n);\n\nconst runtimeInstance = new agentcore.Runtime(this, \"MyAgentRuntime\", {\n  runtimeName: \"myAgent\",\n  agentRuntimeArtifact: agentRuntimeArtifact,\n});\n```\n\n#### Option 4: Use an ECR container image URI\n\nReference an ECR container image directly by its URI. This is useful when you have a pre-existing ECR image URI from CloudFormation parameters or cross-stack references. No IAM permissions are automatically granted - you must ensure the runtime has ECR pull permissions.\n\n```typescript fixture=default\n// Direct URI reference\nconst agentRuntimeArtifact = agentcore.AgentRuntimeArtifact.fromImageUri(\n  \"123456789012.dkr.ecr.us-east-1.amazonaws.com/my-agent:v1.0.0\"\n);\n\nconst runtime = new agentcore.Runtime(this, \"MyAgentRuntime\", {\n  runtimeName: \"myAgent\",\n  agentRuntimeArtifact: agentRuntimeArtifact,\n});\n```\n\nYou can also use CloudFormation parameters or references:\n\n```typescript fixture=default\n// Using a CloudFormation parameter\nconst imageUriParam = new cdk.CfnParameter(this, \"ImageUri\", {\n  type: \"String\",\n  description: \"Container image URI for the agent runtime\",\n});\n\nconst agentRuntimeArtifact = agentcore.AgentRuntimeArtifact.fromImageUri(\n  imageUriParam.valueAsString\n);\n\nconst runtime = new agentcore.Runtime(this, \"MyAgentRuntime\", {\n  runtimeName: \"myAgent\",\n  agentRuntimeArtifact: agentRuntimeArtifact,\n});\n```\n\n### Granting Permissions to Invoke Bedrock Models or Inference Profiles\n\nTo grant the runtime permissions to invoke Bedrock models or inference profiles:\n\n```typescript fixture=default\n// Note: This example uses @aws-cdk/aws-bedrock-alpha which must be installed separately\ndeclare const runtime: agentcore.Runtime;\n\n// Define the Bedrock Foundation Model\nconst model = bedrock.BedrockFoundationModel.ANTHROPIC_CLAUDE_3_7_SONNET_V1_0;\n\n// Grant the runtime permissions to invoke the model\nmodel.grantInvoke(runtime);\n\n// Create a cross-region inference profile for Claude 3.7 Sonnet\nconst inferenceProfile = bedrock.CrossRegionInferenceProfile.fromConfig({\n  geoRegion: bedrock.CrossRegionInferenceProfileRegion.US,\n  model: bedrock.BedrockFoundationModel.ANTHROPIC_CLAUDE_3_7_SONNET_V1_0\n});\n\n// Grant the runtime permissions to invoke the inference profile\ninferenceProfile.grantInvoke(runtime);\n```\n\n### Runtime Versioning\n\nAmazon Bedrock AgentCore automatically manages runtime versioning to ensure safe deployments and rollback capabilities.\nWhen you create an agent runtime, AgentCore automatically creates version 1 (V1). Each subsequent update to the\nruntime configuration (such as updating the container image, modifying network settings, or changing protocol configurations)\ncreates a new immutable version. These versions contain complete, self-contained configurations that can be referenced by endpoints,\nallowing you to maintain different versions for different environments or gradually roll out updates.\n\n#### Managing Endpoints and Versions\n\nAmazon Bedrock AgentCore automatically manages runtime versioning to provide safe deployments and rollback capabilities. You can follow\nthe steps below to understand how to use versioning with runtime for controlled deployments across different environments.\n\n##### Step 1: Initial Deployment\n\nWhen you first create an agent runtime, AgentCore automatically creates Version 1 of your runtime. At this point, a DEFAULT endpoint is\nautomatically created that points to Version 1. This DEFAULT endpoint serves as the main access point for your runtime.\n\n```typescript fixture=default\nconst repository = new ecr.Repository(this, \"TestRepository\", {\n  repositoryName: \"test-agent-runtime\",\n});\n\nconst runtime = new agentcore.Runtime(this, \"MyAgentRuntime\", {\n  runtimeName: \"myAgent\",\n  agentRuntimeArtifact: agentcore.AgentRuntimeArtifact.fromEcrRepository(repository, \"v1.0.0\"),\n});\n```\n\n##### Step 2: Creating Custom Endpoints\n\nAfter the initial deployment, you can create additional endpoints for different environments. For example, you might create a \"production\"\nendpoint that explicitly points to Version 1. This allows you to maintain stable access points for specific environments while keeping the\nflexibility to test newer versions elsewhere.\n\n```typescript fixture=default\nconst repository = new ecr.Repository(this, \"TestRepository\", {\n  repositoryName: \"test-agent-runtime\",\n});\n\nconst runtime = new agentcore.Runtime(this, \"MyAgentRuntime\", {\n  runtimeName: \"myAgent\",\n  agentRuntimeArtifact: agentcore.AgentRuntimeArtifact.fromEcrRepository(repository, \"v1.0.0\"),\n});\n\nconst prodEndpoint = runtime.addEndpoint(\"production\", {\n  version: \"1\",\n  description: \"Stable production endpoint - pinned to v1\"\n});\n```\n\n##### Step 3: Runtime Update Deployment\n\nWhen you update the runtime configuration (such as updating the container image, modifying network settings, or changing protocol\nconfigurations), AgentCore automatically creates a new version (Version 2). Upon this update:\n\n- Version 2 is created automatically with the new configuration\n- The DEFAULT endpoint automatically updates to point to Version 2\n- Any explicitly pinned endpoints (like the production endpoint) remain on their specified versions\n\n```typescript fixture=default\nconst repository = new ecr.Repository(this, \"TestRepository\", {\n  repositoryName: \"test-agent-runtime\",\n});\n\nconst agentRuntimeArtifactNew = agentcore.AgentRuntimeArtifact.fromEcrRepository(repository, \"v2.0.0\");\n\nconst runtime = new agentcore.Runtime(this, \"MyAgentRuntime\", {\n  runtimeName: \"myAgent\",\n  agentRuntimeArtifact: agentRuntimeArtifactNew,\n});\n```\n\n##### Step 4: Testing with Staging Endpoints\n\nOnce Version 2 exists, you can create a staging endpoint that points to the new version. This staging endpoint allows you to test the\nnew version in a controlled environment before promoting it to production. This separation ensures that production traffic continues\nto use the stable version while you validate the new version.\n\n```typescript fixture=default\nconst repository = new ecr.Repository(this, \"TestRepository\", {\n  repositoryName: \"test-agent-runtime\",\n});\n\nconst agentRuntimeArtifactNew = agentcore.AgentRuntimeArtifact.fromEcrRepository(repository, \"v2.0.0\");\n\nconst runtime = new agentcore.Runtime(this, \"MyAgentRuntime\", {\n  runtimeName: \"myAgent\",\n  agentRuntimeArtifact: agentRuntimeArtifactNew,\n});\n\nconst stagingEndpoint = runtime.addEndpoint(\"staging\", {\n  version: \"2\",\n  description: \"Staging environment for testing new version\"\n});\n```\n\n##### Step 5: Promoting to Production\n\nAfter thoroughly testing the new version through the staging endpoint, you can update the production endpoint to point to Version 2.\nThis controlled promotion process ensures that you can validate changes before they affect production traffic.\n\n```typescript fixture=default\nconst repository = new ecr.Repository(this, \"TestRepository\", {\n  repositoryName: \"test-agent-runtime\",\n});\n\nconst agentRuntimeArtifactNew = agentcore.AgentRuntimeArtifact.fromEcrRepository(repository, \"v2.0.0\");\n\nconst runtime = new agentcore.Runtime(this, \"MyAgentRuntime\", {\n  runtimeName: \"myAgent\",\n  agentRuntimeArtifact: agentRuntimeArtifactNew,\n});\n\nconst prodEndpoint = runtime.addEndpoint(\"production\", {\n  version: \"2\",  // New version added here\n  description: \"Stable production endpoint\"\n});\n```\n\n### Creating Standalone Runtime Endpoints\n\nRuntimeEndpoint can also be created as a standalone resource.\n\n#### Example: Creating an endpoint for an existing runtime\n\n```typescript fixture=default\n// Reference an existing runtime by its ID\nconst existingRuntimeId = \"abc123-runtime-id\"; // The ID of an existing runtime\n\n// Create a standalone endpoint\nconst endpoint = new agentcore.RuntimeEndpoint(this, \"MyEndpoint\", {\n  endpointName: \"production\",\n  agentRuntimeId: existingRuntimeId,\n  agentRuntimeVersion: \"1\", // Specify which version to use\n  description: \"Production endpoint for existing runtime\"\n});\n```\n\n### Runtime Authentication Configuration\n\nThe AgentCore Runtime supports multiple authentication modes to secure access to your agent endpoints. Authentication is configured during runtime creation using the `RuntimeAuthorizerConfiguration` class's static factory methods.\n\n#### IAM Authentication (Default)\n\nIAM authentication is the default mode, when no authorizerConfiguration is set then the underlying service use IAM.\n\n#### Cognito Authentication\n\nTo configure AWS Cognito User Pool authentication:\n\n```typescript fixture=default\ndeclare const userPool: cognito.UserPool;\ndeclare const userPoolClient: cognito.UserPoolClient;\ndeclare const anotherUserPoolClient: cognito.UserPoolClient;\n\nconst repository = new ecr.Repository(this, \"TestRepository\", {\n  repositoryName: \"test-agent-runtime\",\n});\nconst agentRuntimeArtifact = agentcore.AgentRuntimeArtifact.fromEcrRepository(repository, \"v1.0.0\");\n\nconst runtime = new agentcore.Runtime(this, \"MyAgentRuntime\", {\n  runtimeName: \"myAgent\",\n  agentRuntimeArtifact: agentRuntimeArtifact,\n  authorizerConfiguration: agentcore.RuntimeAuthorizerConfiguration.usingCognito(\n    userPool, // User Pool (required)\n    [userPoolClient, anotherUserPoolClient], // User Pool Clients\n  ),\n});\n```\n\n#### JWT Authentication\n\nTo configure custom JWT authentication with your own OpenID Connect (OIDC) provider:\n\n```typescript fixture=default\nconst repository = new ecr.Repository(this, \"TestRepository\", {\n  repositoryName: \"test-agent-runtime\",\n});\nconst agentRuntimeArtifact = agentcore.AgentRuntimeArtifact.fromEcrRepository(repository, \"v1.0.0\");\n\nconst runtime = new agentcore.Runtime(this, \"MyAgentRuntime\", {\n  runtimeName: \"myAgent\",\n  agentRuntimeArtifact: agentRuntimeArtifact,\n  authorizerConfiguration: agentcore.RuntimeAuthorizerConfiguration.usingJWT(\n    \"https://example.com/.well-known/openid-configuration\",  // Discovery URL (required)\n    [\"client1\", \"client2\"],  // Allowed Client IDs (optional)\n    [\"audience1\"]           // Allowed Audiences (optional)\n  ),\n});\n```\n\n**Note**: The discovery URL must end with `/.well-known/openid-configuration`.\n\n#### OAuth Authentication\n\nTo configure OAuth 2.0 authentication:\n\n```typescript fixture=default\nconst repository = new ecr.Repository(this, \"TestRepository\", {\n  repositoryName: \"test-agent-runtime\",\n});\nconst agentRuntimeArtifact = agentcore.AgentRuntimeArtifact.fromEcrRepository(repository, \"v1.0.0\");\n\nconst runtime = new agentcore.Runtime(this, \"MyAgentRuntime\", {\n  runtimeName: \"myAgent\",\n  agentRuntimeArtifact: agentRuntimeArtifact,\n  authorizerConfiguration: agentcore.RuntimeAuthorizerConfiguration.usingOAuth(\n    \"https://github.com/.well-known/openid-configuration\",\n    \"oauth_client_123\",\n  ),\n});\n```\n\n#### Using a Custom IAM Role\n\nInstead of using the auto-created execution role, you can provide your own IAM role with specific permissions:\nThe auto-created role includes all necessary baseline permissions for ECR access, CloudWatch logging, and X-Ray tracing. When providing a custom role, ensure these permissions are included.\n\n### Runtime Network Configuration\n\nThe AgentCore Runtime supports two network modes for deployment:\n\n#### Public Network Mode (Default)\n\nBy default, runtimes are deployed in PUBLIC network mode, which provides internet access suitable for less sensitive or open-use scenarios:\n\n```typescript fixture=default\nconst repository = new ecr.Repository(this, \"TestRepository\", {\n  repositoryName: \"test-agent-runtime\",\n});\nconst agentRuntimeArtifact = agentcore.AgentRuntimeArtifact.fromEcrRepository(repository, \"v1.0.0\");\n\n// Explicitly using public network (this is the default)\nconst runtime = new agentcore.Runtime(this, \"MyAgentRuntime\", {\n  runtimeName: \"myAgent\",\n  agentRuntimeArtifact: agentRuntimeArtifact,\n  networkConfiguration: agentcore.RuntimeNetworkConfiguration.usingPublicNetwork(),\n});\n```\n\n#### VPC Network Mode\n\nFor enhanced security and network isolation, you can deploy your runtime within a VPC:\n\n```typescript fixture=default\nconst repository = new ecr.Repository(this, \"TestRepository\", {\n  repositoryName: \"test-agent-runtime\",\n});\nconst agentRuntimeArtifact = agentcore.AgentRuntimeArtifact.fromEcrRepository(repository, \"v1.0.0\");\n\n// Create or use an existing VPC\nconst vpc = new ec2.Vpc(this, 'MyVpc', {\n  maxAzs: 2,\n});\n\n// Configure runtime with VPC\nconst runtime = new agentcore.Runtime(this, \"MyAgentRuntime\", {\n  runtimeName: \"myAgent\",\n  agentRuntimeArtifact: agentRuntimeArtifact,\n  networkConfiguration: agentcore.RuntimeNetworkConfiguration.usingVpc(this, {\n    vpc: vpc,\n    vpcSubnets: { subnetType: ec2.SubnetType.PRIVATE_WITH_EGRESS },\n    // Optionally specify security groups, or one will be created automatically\n    // securityGroups: [mySecurityGroup],\n  }),\n});\n\n```\n\n#### Managing Security Groups with VPC Configuration\n\nWhen using VPC mode, the Runtime implements `ec2.IConnectable`, allowing you to manage network access using the `connections` property:\n\n```typescript fixture=default\nconst vpc = new ec2.Vpc(this, 'MyVpc', {\n  maxAzs: 2,\n});\n\nconst repository = new ecr.Repository(this, \"TestRepository\", {\n  repositoryName: \"test-agent-runtime\",\n});\nconst agentRuntimeArtifact = agentcore.AgentRuntimeArtifact.fromEcrRepository(repository, \"v1.0.0\");\n\n// Create runtime with VPC configuration\nconst runtime = new agentcore.Runtime(this, \"MyAgentRuntime\", {\n  runtimeName: \"myAgent\",\n  agentRuntimeArtifact: agentRuntimeArtifact,\n  networkConfiguration: agentcore.RuntimeNetworkConfiguration.usingVpc(this, {\n    vpc: vpc,\n    vpcSubnets: { subnetType: ec2.SubnetType.PRIVATE_WITH_EGRESS },\n  }),\n});\n\n// Now you can manage network access using the connections property\n// Allow inbound HTTPS traffic from a specific security group\nconst webServerSecurityGroup = new ec2.SecurityGroup(this, 'WebServerSG', { vpc });\nruntime.connections.allowFrom(webServerSecurityGroup, ec2.Port.tcp(443), 'Allow HTTPS from web servers');\n\n// Allow outbound connections to a database\nconst databaseSecurityGroup = new ec2.SecurityGroup(this, 'DatabaseSG', { vpc });\nruntime.connections.allowTo(databaseSecurityGroup, ec2.Port.tcp(5432), 'Allow PostgreSQL connection');\n\n// Allow outbound HTTPS to anywhere (for external API calls)\nruntime.connections.allowToAnyIpv4(ec2.Port.tcp(443), 'Allow HTTPS outbound');\n```\n\n### Runtime IAM Permissions\n\nThe Runtime construct provides convenient methods for granting IAM permissions to principals that need to invoke the runtime or manage its execution role.\n\n```typescript fixture=default\nconst repository = new ecr.Repository(this, \"TestRepository\", {\n  repositoryName: \"test-agent-runtime\",\n});\nconst agentRuntimeArtifact = agentcore.AgentRuntimeArtifact.fromEcrRepository(repository, \"v1.0.0\");\n\n// Create a runtime\nconst runtime = new agentcore.Runtime(this, \"MyRuntime\", {\n  runtimeName: \"my_runtime\",\n  agentRuntimeArtifact: agentRuntimeArtifact,\n});\n\n// Create a Lambda function that needs to invoke the runtime\nconst invokerFunction = new lambda.Function(this, \"InvokerFunction\", {\n  runtime: lambda.Runtime.PYTHON_3_12,\n  handler: \"index.handler\",\n  code: lambda.Code.fromInline(`\nimport boto3\ndef handler(event, context):\n    client = boto3.client('bedrock-agentcore')\n    # Invoke the runtime...\n  `),\n});\n\n// Grant permission to invoke the runtime directly\nruntime.grantInvokeRuntime(invokerFunction);\n\n// Grant permission to invoke the runtime on behalf of a user\n// (requires X-Amzn-Bedrock-AgentCore-Runtime-User-Id header)\nruntime.grantInvokeRuntimeForUser(invokerFunction);\n\n// Grant both invoke permissions (most common use case)\nruntime.grantInvoke(invokerFunction);\n\n// Grant specific custom permissions to the runtime's execution role\nruntime.grant(['bedrock:InvokeModel'], ['arn:aws:bedrock:*:*:*']);\n\n// Add a policy statement to the runtime's execution role\nruntime.addToRolePolicy(new iam.PolicyStatement({\n  actions: ['s3:GetObject'],\n  resources: ['arn:aws:s3:::my-bucket/*'],\n}));\n```\n\n### Other configuration\n\n#### Lifecycle configuration\n\nThe LifecycleConfiguration input parameter to CreateAgentRuntime lets you manage the lifecycle of runtime sessions and resources in Amazon Bedrock AgentCore Runtime. This configuration helps optimize resource utilization by automatically cleaning up idle sessions and preventing long-running instances from consuming resources indefinitely.\n\nYou can configure:\n\n- idleRuntimeSessionTimeout: Timeout in seconds for idle runtime sessions. When a session remains idle for this duration, it will trigger termination. Termination can last up to 15 seconds due to logging and other process completion. Default: 900 seconds (15 minutes)\n- maxLifetime: Maximum lifetime for the instance in seconds. Once reached, instances will initialize termination. Termination can last up to 15 seconds due to logging and other process completion. Default: 28800 seconds (8 hours)\n\nFor additional information, please refer to the [documentation](https://docs.aws.amazon.com/bedrock-agentcore/latest/devguide/runtime-lifecycle-settings.html).\n\n```typescript fixture=default\nconst repository = new ecr.Repository(this, \"TestRepository\", {\n  repositoryName: \"test-agent-runtime\",\n});\n\nconst agentRuntimeArtifact = agentcore.AgentRuntimeArtifact.fromEcrRepository(repository, \"v1.0.0\");\n\nnew agentcore.Runtime(this, 'test-runtime', {\n  runtimeName: 'test_runtime',\n  agentRuntimeArtifact: agentRuntimeArtifact,\n  lifecycleConfiguration: {\n    idleRuntimeSessionTimeout: Duration.minutes(10),\n    maxLifetime: Duration.hours(4),\n  },\n});\n```\n\n#### Request header configuration\n\nCustom headers let you pass contextual information from your application directly to your agent code without cluttering the main request payload. This includes authentication tokens like JWT (JSON Web Tokens, which contain user identity and authorization claims) through the Authorization header, allowing your agent to make decisions based on who is calling it. You can also pass custom metadata like user preferences, session identifiers, or trace context using headers prefixed with X-Amzn-Bedrock-AgentCore-Runtime-Custom-, giving your agent access to up to 20 pieces of runtime context that travel alongside each request. This information can be also used in downstream systems like AgentCore Memory that you can namespace based on those characteristics like user_id or aud in claims like line of business.\n\nFor additional information, please refer to the [documentation](https://docs.aws.amazon.com/bedrock-agentcore/latest/devguide/runtime-header-allowlist.html).\n\n```typescript fixture=default\nconst repository = new ecr.Repository(this, \"TestRepository\", {\n  repositoryName: \"test-agent-runtime\",\n});\n\nconst agentRuntimeArtifact = agentcore.AgentRuntimeArtifact.fromEcrRepository(repository, \"v1.0.0\");\n\nnew agentcore.Runtime(this, 'test-runtime', {\n  runtimeName: 'test_runtime',\n  agentRuntimeArtifact: agentRuntimeArtifact,\n  requestHeaderConfiguration: {\n    allowlistedHeaders: ['X-Amzn-Bedrock-AgentCore-Runtime-Custom-H1'],\n  },\n});\n```\n\n## Browser\n\nThe Amazon Bedrock AgentCore Browser provides a secure, cloud-based browser that enables AI agents to interact with websites. It includes security features such as session isolation, built-in observability through live viewing, CloudTrail logging, and session replay capabilities.\n\nAdditional information about the browser tool can be found in the [official documentation](https://docs.aws.amazon.com/bedrock-agentcore/latest/devguide/browser-tool.html)\n\n### Browser Network modes\n\nThe Browser construct supports the following network modes:\n\n1. **Public Network Mode** (`BrowserNetworkMode.usingPublicNetwork()`) - Default\n\n   - Allows internet access for web browsing and external API calls\n   - Suitable for scenarios where agents need to interact with publicly available websites\n   - Enables full web browsing capabilities\n   - VPC mode is not supported with this option\n\n2. **VPC (Virtual Private Cloud)** (`BrowserNetworkMode.usingVpc()`)\n\n   - Select whether to run the browser in a virtual private cloud (VPC).\n   - By configuring VPC connectivity, you enable secure access to private resources such as databases, internal APIs, and services within your VPC.\n\n    While the VPC itself is mandatory, these are optional:\n    - Subnets - if not provided, CDK will select appropriate subnets from the VPC\n    - Security Groups - if not provided, CDK will create a default security group\n    - Specific subnet selection criteria - you can let CDK choose automatically\n\nFor more information on VPC connectivity for Amazon Bedrock AgentCore Browser, please refer to the [official documentation](https://docs.aws.amazon.com/bedrock-agentcore/latest/devguide/agentcore-vpc.html).\n\n### Browser Properties\n\n| Name | Type | Required | Description |\n|------|------|----------|-------------|\n| `browserCustomName` | `string` | Yes | The name of the browser. Must start with a letter and can be up to 48 characters long. Pattern: `[a-zA-Z][a-zA-Z0-9_]{0,47}` |\n| `description` | `string` | No | Optional description for the browser. Can have up to 200 characters |\n| `networkConfiguration` | `BrowserNetworkConfiguration` | No | Network configuration for browser. Defaults to PUBLIC network mode |\n| `recordingConfig` | `RecordingConfig` | No | Recording configuration for browser. Defaults to no recording |\n| `executionRole` | `iam.IRole` | No | The IAM role that provides permissions for the browser to access AWS services. A new role will be created if not provided |\n| `tags` | `{ [key: string]: string }` | No | Tags to apply to the browser resource |\n| `browserSigning` | BrowserSigning | No | Browser signing configuration. Defaults to DISABLED |\n\n### Basic Browser Creation\n\n```typescript fixture=default\n// Create a basic browser with public network access\nconst browser = new agentcore.BrowserCustom(this, \"MyBrowser\", {\n  browserCustomName: \"my_browser\",\n  description: \"A browser for web automation\",\n});\n```\n\n### Browser with Tags\n\n```typescript fixture=default\n// Create a browser with custom tags\nconst browser = new agentcore.BrowserCustom(this, \"MyBrowser\", {\n  browserCustomName: \"my_browser\",\n  description: \"A browser for web automation with tags\",\n  networkConfiguration: agentcore.BrowserNetworkConfiguration.usingPublicNetwork(),\n  tags: {\n    Environment: \"Production\",\n    Team: \"AI/ML\",\n    Project: \"AgentCore\",\n  },\n});\n```\n\n### Browser with VPC\n\n```typescript fixture=default\nconst browser = new agentcore.BrowserCustom(this, 'BrowserVpcWithRecording', {\n  browserCustomName: 'browser_recording',\n  networkConfiguration: agentcore.BrowserNetworkConfiguration.usingVpc(this, {\n    vpc: new ec2.Vpc(this, 'VPC', { restrictDefaultSecurityGroup: false }),\n  }),\n});\n```\n\nBrowser exposes a [connections](https://docs.aws.amazon.com/cdk/api/v2/docs/aws-cdk-lib.aws_ec2.Connections.html) property. This property returns a connections object, which simplifies the process of defining and managing ingress and egress rules for security groups in your AWS CDK applications. Instead of directly manipulating security group rules, you interact with the Connections object of a construct, which then translates your connectivity requirements into the appropriate security group rules. For instance:\n\n```typescript fixture=default\nconst vpc = new ec2.Vpc(this, 'testVPC');\n\nconst browser = new agentcore.BrowserCustom(this, 'test-browser', {\n  browserCustomName: 'test_browser',\n  networkConfiguration: agentcore.BrowserNetworkConfiguration.usingVpc(this, {\n    vpc: vpc,\n  }),\n});\n\nbrowser.connections.addSecurityGroup(new ec2.SecurityGroup(this, 'AdditionalGroup', { vpc }));\n```\n\nSo security groups can be added after the browser construct creation. You can use methods like allowFrom() and allowTo() to grant ingress access to/egress access from a specified peer over a given portRange. The Connections object automatically adds the necessary ingress or egress rules to the security group(s) associated with the calling construct.\n\n### Browser with Recording Configuration\n\n```typescript fixture=default\n// Create an S3 bucket for recordings\nconst recordingBucket = new s3.Bucket(this, \"RecordingBucket\", {\n  bucketName: \"my-browser-recordings\",\n  removalPolicy: RemovalPolicy.DESTROY, // For demo purposes\n});\n\n// Create browser with recording enabled\nconst browser = new agentcore.BrowserCustom(this, \"MyBrowser\", {\n  browserCustomName: \"my_browser\",\n  description: \"Browser with recording enabled\",\n  networkConfiguration: agentcore.BrowserNetworkConfiguration.usingPublicNetwork(),\n  recordingConfig: {\n    enabled: true,\n    s3Location: {\n      bucketName: recordingBucket.bucketName,\n      objectKey: \"browser-recordings/\",\n    },\n  },\n});\n```\n\n### Browser with Custom Execution Role\n\n```typescript fixture=default\n// Create a custom execution role\nconst executionRole = new iam.Role(this, \"BrowserExecutionRole\", {\n  assumedBy: new iam.ServicePrincipal(\"bedrock-agentcore.amazonaws.com\"),\n  managedPolicies: [\n    iam.ManagedPolicy.fromAwsManagedPolicyName(\"AmazonBedrockAgentCoreBrowserExecutionRolePolicy\"),\n  ],\n});\n\n// Create browser with custom execution role\nconst browser = new agentcore.BrowserCustom(this, \"MyBrowser\", {\n  browserCustomName: \"my_browser\",\n  description: \"Browser with custom execution role\",\n  networkConfiguration: agentcore.BrowserNetworkConfiguration.usingPublicNetwork(),\n  executionRole: executionRole,\n});\n```\n\n### Browser with S3 Recording and Permissions\n\n```typescript fixture=default\n// Create an S3 bucket for recordings\nconst recordingBucket = new s3.Bucket(this, \"RecordingBucket\", {\n  bucketName: \"my-browser-recordings\",\n  removalPolicy: RemovalPolicy.DESTROY, // For demo purposes\n});\n\n// Create browser with recording enabled\nconst browser = new agentcore.BrowserCustom(this, \"MyBrowser\", {\n  browserCustomName: \"my_browser\",\n  description: \"Browser with recording enabled\",\n  networkConfiguration: agentcore.BrowserNetworkConfiguration.usingPublicNetwork(),\n  recordingConfig: {\n    enabled: true,\n    s3Location: {\n      bucketName: recordingBucket.bucketName,\n      objectKey: \"browser-recordings/\",\n    },\n  },\n});\n\n// The browser construct automatically grants S3 permissions to the execution role\n// when recording is enabled, so no additional IAM configuration is needed\n```\n\n### Browser with Browser signing\n\nAI agents need to browse the web on your behalf. When your agent visits a website to gather information, complete a form, or verify data, it encounters the same defenses designed to stop unwanted bots: CAPTCHAs, rate limits, and outright blocks.\n\nAmazon Bedrock AgentCore Browser supports Web Bot Auth. Web Bot Auth is a draft IETF protocol that gives agents verifiable cryptographic identities. When you enable Web Bot Auth in AgentCore Browser, the service issues cryptographic credentials that websites can verify. The agent presents these credentials with every request. The WAF may now additionally check the signature, confirm it matches a trusted directory, and allow the request through if verified bots are allowed by the domain owner and other WAF checks are clear.\n\nTo enable the browser to sign requests using the Web Bot Auth protocol, create a browser tool with the browserSigning configuration:\n\n```typescript fixture=default\nconst browser = new agentcore.BrowserCustom(this, 'test-browser', {\n  browserCustomName: 'test_browser',\n  browserSigning: agentcore.BrowserSigning.ENABLED\n});\n```\n\n### Browser IAM Permissions\n\nThe Browser construct provides convenient methods for granting IAM permissions:\n\n```typescript fixture=default\n// Create a browser\nconst browser = new agentcore.BrowserCustom(this, \"MyBrowser\", {\n  browserCustomName: \"my_browser\",\n  description: \"Browser for web automation\",\n  networkConfiguration: agentcore.BrowserNetworkConfiguration.usingPublicNetwork(),\n});\n\n// Create a role that needs access to the browser\nconst userRole = new iam.Role(this, \"UserRole\", {\n  assumedBy: new iam.ServicePrincipal(\"lambda.amazonaws.com\"),\n});\n\n// Grant read permissions (Get and List actions)\nbrowser.grantRead(userRole);\n\n// Grant use permissions (Start, Update, Stop actions)\nbrowser.grantUse(userRole);\n\n// Grant specific custom permissions\nbrowser.grant(userRole, \"bedrock-agentcore:GetBrowserSession\");\n```\n\n## Code Interpreter\n\nThe Amazon Bedrock AgentCore Code Interpreter enables AI agents to write and execute code securely in sandbox environments, enhancing their accuracy and expanding their ability to solve complex end-to-end tasks. This is critical in Agentic AI applications where the agents may execute arbitrary code that can lead to data compromise or security risks. The AgentCore Code Interpreter tool provides secure code execution, which helps you avoid running into these issues.\n\nFor more information about code interpreter, please refer to the [official documentation](https://docs.aws.amazon.com/bedrock-agentcore/latest/devguide/code-interpreter-tool.html)\n\n### Code Interpreter Network Modes\n\nThe Code Interpreter construct supports the following network modes:\n\n1. **Public Network Mode** (`CodeInterpreterNetworkMode.usingPublicNetwork()`) - Default\n\n   - Allows internet access for package installation and external API calls\n   - Suitable for development and testing environments\n   - Enables downloading Python packages from PyPI\n\n2. **Sandbox Network Mode** (`CodeInterpreterNetworkMode.usingSandboxNetwork()`)\n   - Isolated network environment with no internet access\n   - Suitable for production environments with strict security requirements\n   - Only allows access to pre-installed packages and local resources\n\n3. **VPC (Virtual Private Cloud)** (`CodeInterpreterNetworkMode.usingVpc()`)\n   - Select whether to run the browser in a virtual private cloud (VPC).\n   - By configuring VPC connectivity, you enable secure access to private resources such as databases, internal APIs, and services within your VPC.\n\n    While the VPC itself is mandatory, these are optional:\n    - Subnets - if not provided, CDK will select appropriate subnets from the VPC\n    - Security Groups - if not provided, CDK will create a default security group\n    - Specific subnet selection criteria - you can let CDK choose automatically\n\nFor more information on VPC connectivity for Amazon Bedrock AgentCore Browser, please refer to the [official documentation](https://docs.aws.amazon.com/bedrock-agentcore/latest/devguide/agentcore-vpc.html).\n\n### Code Interpreter Properties\n\n| Name | Type | Required | Description |\n|------|------|----------|-------------|\n| `codeInterpreterCustomName` | `string` | Yes | The name of the code interpreter. Must start with a letter and can be up to 48 characters long. Pattern: `[a-zA-Z][a-zA-Z0-9_]{0,47}` |\n| `description` | `string` | No | Optional description for the code interpreter. Can have up to 200 characters |\n| `executionRole` | `iam.IRole` | No | The IAM role that provides permissions for the code interpreter to access AWS services. A new role will be created if not provided |\n| `networkConfiguration` | `CodeInterpreterNetworkConfiguration` | No | Network configuration for code interpreter. Defaults to PUBLIC network mode |\n| `tags` | `{ [key: string]: string }` | No | Tags to apply to the code interpreter resource |\n\n### Basic Code Interpreter Creation\n\n```typescript fixture=default\n// Create a basic code interpreter with public network access\nconst codeInterpreter = new agentcore.CodeInterpreterCustom(this, \"MyCodeInterpreter\", {\n  codeInterpreterCustomName: \"my_code_interpreter\",\n  description: \"A code interpreter for Python execution\",\n});\n```\n\n### Code Interpreter with VPC\n\n```typescript fixture=default\nconst codeInterpreter = new agentcore.CodeInterpreterCustom(this, \"MyCodeInterpreter\", {\n  codeInterpreterCustomName: \"my_sandbox_interpreter\",\n  description: \"Code interpreter with isolated network access\",\n  networkConfiguration: agentcore.BrowserNetworkConfiguration.usingVpc(this, {\n    vpc: new ec2.Vpc(this, 'VPC', { restrictDefaultSecurityGroup: false }),\n  }),\n});\n```\n\nCode Interpreter exposes a [connections](https://docs.aws.amazon.com/cdk/api/v2/docs/aws-cdk-lib.aws_ec2.Connections.html) property. This property returns a connections object, which simplifies the process of defining and managing ingress and egress rules for security groups in your AWS CDK applications. Instead of directly manipulating security group rules, you interact with the Connections object of a construct, which then translates your connectivity requirements into the appropriate security group rules. For instance:\n\n```typescript fixture=default\nconst vpc = new ec2.Vpc(this, 'testVPC');\n\nconst codeInterpreter = new agentcore.CodeInterpreterCustom(this, \"MyCodeInterpreter\", {\n  codeInterpreterCustomName: \"my_sandbox_interpreter\",\n  description: \"Code interpreter with isolated network access\",\n  networkConfiguration: agentcore.BrowserNetworkConfiguration.usingVpc(this, {\n    vpc: vpc,\n  }),\n});\n\ncodeInterpreter.connections.addSecurityGroup(new ec2.SecurityGroup(this, 'AdditionalGroup', { vpc }));\n```\n\nSo security groups can be added after the browser construct creation. You can use methods like allowFrom() and allowTo() to grant ingress access to/egress access from a specified peer over a given portRange. The Connections object automatically adds the necessary ingress or egress rules to the security group(s) associated with the calling construct.\n\n### Code Interpreter with Sandbox Network Mode\n\n```typescript fixture=default\n// Create code interpreter with sandbox network mode (isolated)\nconst codeInterpreter = new agentcore.CodeInterpreterCustom(this, \"MyCodeInterpreter\", {\n  codeInterpreterCustomName: \"my_sandbox_interpreter\",\n  description: \"Code interpreter with isolated network access\",\n  networkConfiguration: agentcore.CodeInterpreterNetworkConfiguration.usingSandboxNetwork(),\n});\n```\n\n### Code Interpreter with Custom Execution Role\n\n```typescript fixture=default\n// Create a custom execution role\nconst executionRole = new iam.Role(this, \"CodeInterpreterExecutionRole\", {\n  assumedBy: new iam.ServicePrincipal(\"bedrock-agentcore.amazonaws.com\"),\n});\n\n// Create code interpreter with custom execution role\nconst codeInterpreter = new agentcore.CodeInterpreterCustom(this, \"MyCodeInterpreter\", {\n  codeInterpreterCustomName: \"my_code_interpreter\",\n  description: \"Code interpreter with custom execution role\",\n  networkConfiguration: agentcore.CodeInterpreterNetworkConfiguration.usingPublicNetwork(),\n  executionRole: executionRole,\n});\n```\n\n### Code Interpreter IAM Permissions\n\nThe Code Interpreter construct provides convenient methods for granting IAM permissions:\n\n```typescript fixture=default\n// Create a code interpreter\nconst codeInterpreter = new agentcore.CodeInterpreterCustom(this, \"MyCodeInterpreter\", {\n  codeInterpreterCustomName: \"my_code_interpreter\",\n  description: \"Code interpreter for Python execution\",\n  networkConfiguration: agentcore.CodeInterpreterNetworkConfiguration.usingPublicNetwork(),\n});\n\n// Create a role that needs access to the code interpreter\nconst userRole = new iam.Role(this, \"UserRole\", {\n  assumedBy: new iam.ServicePrincipal(\"lambda.amazonaws.com\"),\n});\n\n// Grant read permissions (Get and List actions)\ncodeInterpreter.grantRead(userRole);\n\n// Grant use permissions (Start, Invoke, Stop actions)\ncodeInterpreter.grantUse(userRole);\n\n// Grant specific custom permissions\ncodeInterpreter.grant(userRole, \"bedrock-agentcore:GetCodeInterpreterSession\");\n```\n\n### Code interpreter with tags\n\n```typescript fixture=default\n// Create code interpreter with sandbox network mode (isolated)\nconst codeInterpreter = new agentcore.CodeInterpreterCustom(this, \"MyCodeInterpreter\", {\n  codeInterpreterCustomName: \"my_sandbox_interpreter\",\n  description: \"Code interpreter with isolated network access\",\n  networkConfiguration: agentcore.CodeInterpreterNetworkConfiguration.usingPublicNetwork(),\n  tags: {\n    Environment: \"Production\",\n    Team: \"AI/ML\",\n    Project: \"AgentCore\",\n  },\n});\n```\n\n## Gateway\n\nThe Gateway construct provides a way to create Amazon Bedrock Agent Core Gateways, which serve as integration points between agents and external services.\n\n### Gateway Properties\n\n| Name | Type | Required | Description |\n|------|------|----------|-------------|\n| `gatewayName` | `string` | Yes | The name of the gateway. Valid characters are a-z, A-Z, 0-9, _ (underscore) and - (hyphen). Maximum 100 characters |\n| `description` | `string` | No | Optional description for the gateway. Maximum 200 characters |\n| `protocolConfiguration` | `IGatewayProtocolConfig` | No | The protocol configuration for the gateway. Defaults to MCP protocol |\n| `authorizerConfiguration` | `IGatewayAuthorizerConfig` | No | The authorizer configuration for the gateway. Defaults to Cognito |\n| `exceptionLevel` | `GatewayExceptionLevel` | No | The verbosity of exception messages. Use DEBUG mode to see granular exception messages |\n| `kmsKey` | `kms.IKey` | No | The AWS KMS key used to encrypt data associated with the gateway |\n| `role` | `iam.IRole` | No | The IAM role that provides permissions for the gateway to access AWS services. A new role will be created if not provided |\n| `tags` | `{ [key: string]: string }` | No | Tags for the gateway. A list of key:value pairs of tags to apply to this Gateway resource |\n\n### Basic Gateway Creation\n\nThe protocol configuration defaults to MCP and the inbound auth configuration uses Cognito (it is automatically created on your behalf).\n\n```typescript fixture=default\n// Create a basic gateway with default MCP protocol and Cognito authorizer\nconst gateway = new agentcore.Gateway(this, \"MyGateway\", {\n  gatewayName: \"my-gateway\",\n});\n```\n\n### Protocol configuration\n\nCurrently MCP is the only protocol available. To configure it, use the `protocol` property with `McpProtocolConfiguration`:\n\n- Instructions: Guidance for how to use the gateway with your tools\n- Semantic search: Smart tool discovery that finds the right tools without typical limits. It improves accuracy by finding relevant tools based on context\n- Supported versions: Which MCP protocol versions the gateway can use\n\n```typescript fixture=default\nconst gateway = new agentcore.Gateway(this, \"MyGateway\", {\n  gatewayName: \"my-gateway\",\n  protocolConfiguration: new agentcore.McpProtocolConfiguration({\n    instructions: \"Use this gateway to connect to external MCP tools\",\n    searchType: agentcore.McpGatewaySearchType.SEMANTIC,\n    supportedVersions: [agentcore.MCPProtocolVersion.MCP_2025_03_26],\n  }),\n});\n```\n\n### Inbound authorization\n\nBefore you create your gateway, you must set up inbound authorization. Inbound authorization validates users who attempt to access targets through\nyour AgentCore gateway. By default, if not provided, the construct will create and configure Cognito as the default identity provider\n(inbound Auth setup). AgentCore supports the following types of inbound authorization:\n\n**JSON Web Token (JWT)** – A secure and compact token used for authorization. After creating the JWT, you specify it as the authorization\nconfiguration when you create the gateway. You can create a JWT with any of the identity providers at Provider setup and configuration.\n\nYou can configure a custom authorization provider using the `inboundAuthorizer` property with `GatewayAuthorizer.usingCustomJwt()`.\nYou need to specify an OAuth discovery server and client IDs/audiences when you create the gateway. You can specify the following:\n\n- Discovery Url — String that must match the pattern ^.+/\\.well-known/openid-configuration$ for OpenID Connect discovery URLs\n- At least one of the below options depending on the chosen identity provider.\n- Allowed audiences — List of allowed audiences for JWT tokens\n- Allowed clients — List of allowed client identifiers\n\n```typescript fixture=default\nconst gateway = new agentcore.Gateway(this, \"MyGateway\", {\n  gatewayName: \"my-gateway\",\n  authorizerConfiguration: agentcore.GatewayAuthorizer.usingCustomJwt({\n    discoveryUrl: \"https://auth.example.com/.well-known/openid-configuration\",\n    allowedAudience: [\"my-app\"],\n    allowedClients: [\"my-client-id\"],\n  }),\n});\n```\n\n**IAM** – Authorizes through the credentials of the AWS IAM identity trying to access the gateway.\n\n```typescript fixture=default\nconst gateway = new agentcore.Gateway(this, \"MyGateway\", {\n  gatewayName: \"my-gateway\",\n  authorizerConfiguration: agentcore.GatewayAuthorizer.usingAwsIam(),\n});\n\n// Grant access to a Lambda function's role\nconst lambdaRole = new iam.Role(this, \"LambdaRole\", {\n  assumedBy: new iam.ServicePrincipal(\"lambda.amazonaws.com\"),\n});\n\n// The Lambda needs permission to invoke the gateway\ngateway.grantInvoke(lambdaRole);\n```\n\n### Gateway with KMS Encryption\n\nYou can provide a KMS key, and configure the authorizer as well as the protocol configuration.\n\n```typescript fixture=default\n// Create a KMS key for encryption\nconst encryptionKey = new kms.Key(this, \"GatewayEncryptionKey\", {\n  enableKeyRotation: true,\n  description: \"KMS key for gateway encryption\",\n});\n\n// Create gateway with KMS encryption\nconst gateway = new agentcore.Gateway(this, \"MyGateway\", {\n  gatewayName: \"my-encrypted-gateway\",\n  description: \"Gateway with KMS encryption\",\n  protocolConfiguration: new agentcore.McpProtocolConfiguration({\n    instructions: \"Use this gateway to connect to external MCP tools\",\n    searchType: agentcore.McpGatewaySearchType.SEMANTIC,\n    supportedVersions: [agentcore.MCPProtocolVersion.MCP_2025_03_26],\n  }),\n  authorizerConfiguration: agentcore.GatewayAuthorizer.usingCustomJwt({\n    discoveryUrl: \"https://auth.example.com/.well-known/openid-configuration\",\n    allowedAudience: [\"my-app\"],\n    allowedClients: [\"my-client-id\"],\n  }),\n  kmsKey: encryptionKey,\n  exceptionLevel: agentcore.GatewayExceptionLevel.DEBUG,\n});\n```\n\n### Gateway with Custom Execution Role\n\n```typescript fixture=default\n// Create a custom execution role\nconst executionRole = new iam.Role(this, \"GatewayExecutionRole\", {\n  assumedBy: new iam.ServicePrincipal(\"bedrock-agentcore.amazonaws.com\"),\n  managedPolicies: [\n    iam.ManagedPolicy.fromAwsManagedPolicyName(\"AmazonBedrockAgentCoreGatewayExecutionRolePolicy\"),\n  ],\n});\n\n// Create gateway with custom execution role\nconst gateway = new agentcore.Gateway(this, \"MyGateway\", {\n  gatewayName: \"my-gateway\",\n  description: \"Gateway with custom execution role\",\n  protocolConfiguration: new agentcore.McpProtocolConfiguration({\n    instructions: \"Use this gateway to connect to external MCP tools\",\n    searchType: agentcore.McpGatewaySearchType.SEMANTIC,\n    supportedVersions: [agentcore.MCPProtocolVersion.MCP_2025_03_26],\n  }),\n  authorizerConfiguration: agentcore.GatewayAuthorizer.usingCustomJwt({\n    discoveryUrl: \"https://auth.example.com/.well-known/openid-configuration\",\n    allowedAudience: [\"my-app\"],\n    allowedClients: [\"my-client-id\"],\n  }),\n  role: executionRole,\n});\n```\n\n### Gateway IAM Permissions\n\nThe Gateway construct provides convenient methods for granting IAM permissions:\n\n```typescript fixture=default\n// Create a gateway\nconst gateway = new agentcore.Gateway(this, \"MyGateway\", {\n  gatewayName: \"my-gateway\",\n  description: \"Gateway for external service integration\",\n  protocolConfiguration: new agentcore.McpProtocolConfiguration({\n    instructions: \"Use this gateway to connect to external MCP tools\",\n    searchType: agentcore.McpGatewaySearchType.SEMANTIC,\n    supportedVersions: [agentcore.MCPProtocolVersion.MCP_2025_03_26],\n  }),\n  authorizerConfiguration: agentcore.GatewayAuthorizer.usingCustomJwt({\n    discoveryUrl: \"https://auth.example.com/.well-known/openid-configuration\",\n    allowedAudience: [\"my-app\"],\n    allowedClients: [\"my-client-id\"],\n  }),\n});\n\n// Create a role that needs access to the gateway\nconst userRole = new iam.Role(this, \"UserRole\", {\n  assumedBy: new iam.ServicePrincipal(\"lambda.amazonaws.com\"),\n});\n\n// Grant read permissions (Get and List actions)\ngateway.grantRead(userRole);\n\n// Grant manage permissions (Create, Update, Delete actions)\ngateway.grantManage(userRole);\n\n// Grant specific custom permissions\ngateway.grant(userRole, \"bedrock-agentcore:GetGateway\");\n```\n\n## Gateway Target\n\nAfter Creating gateways, you can add targets which define the tools that your gateway will host. Gateway supports multiple target\ntypes including Lambda functions and API specifications (either OpenAPI schemas or Smithy models). Gateway allows you to attach multiple\ntargets to a Gateway and you can change the targets / tools attached to a gateway at any point. Each target can have its own\ncredential provider attached enabling you to securely access targets whether they need IAM, API Key, or OAuth credentials.\n\n### Gateway Target Properties\n\n| Name | Type | Required | Description |\n|------|------|----------|-------------|\n| `gatewayTargetName` | `string` | Yes | The name of the gateway target. Valid characters are a-z, A-Z, 0-9, _ (underscore) and - (hyphen) |\n| `description` | `string` | No | Optional description for the gateway target. Maximum 200 characters |\n| `gateway` | `IGateway` | Yes | The gateway this target belongs to |\n| `targetConfiguration` | `ITargetConfiguration` | Yes | The target configuration (Lambda, OpenAPI, or Smithy). **Note:** Users typically don't create this directly. When using convenience methods like `GatewayTarget.forLambda()`, `GatewayTarget.forOpenApi()`, `GatewayTarget.forSmithy()` or the gateway's `addLambdaTarget()`, `addOpenApiTarget()`, `addSmithyTarget()` methods, this configuration is created internally for you. Only needed when using the GatewayTarget constructor directly for [advanced scenarios](#advanced-usage-direct-configuration-for-gateway-target). |\n| `credentialProviderConfigurations` | `IGatewayCredentialProvider[]` | No | Credential providers for authentication. Defaults to `[GatewayCredentialProvider.fromIamRole()]`. Use `GatewayCredentialProvider.fromApiKeyIdentityArn()`, `GatewayCredentialProvider.fromOauthIdentityArn()`, or `GatewayCredentialProvider.fromIamRole()` |\n| `validateOpenApiSchema` | `boolean` | No | (OpenAPI targets only) Whether to validate the OpenAPI schema at synthesis time. Defaults to `true`. Only applies to inline and local asset schemas. For more information refer here <https://docs.aws.amazon.com/bedrock-agentcore/latest/devguide/gateway-schema-openapi.html> |\n\nThis approach gives you full control over the configuration but is typically not necessary for most use cases.\n\n### Targets types\n\nYou can create the following targets types:\n\n**Lambda Target**: Lambda targets allow you to connect your gateway to AWS Lambda functions that implement your tools. This is useful\nwhen you want to execute custom code in response to tool invocations.\n\n- Supports GATEWAY_IAM_ROLE credential provider only\n- Ideal for custom serverless function integration\n- Need tool schema (tool schema is a blueprint that describes the functions your Lambda provides to AI agents).\n  The construct provide [3 ways to upload a tool schema to Lambda target](#tools-schema-for-lambda-target)\n- When using the default IAM authentication (no `credentialProviderConfigurations` specified),\n  the construct automatocally grants the gateway role permission to invoke your Lambda function (`lambda:InvokeFunction`).\n\n**OpenAPI Schema Target** : OpenAPI widely used standard for describing RESTful APIs. Gateway supports OpenAPI 3.0\nspecifications for defining API targets. It connects to REST APIs using OpenAPI specifications\n\n- Supports OAUTH and API_KEY credential providers (Do not support IAM, you must provide `credentialProviderConfigurations`)\n- Ideal for integrating with external REST services\n- Need API schema. The construct provide [3 ways to upload a API schema to OpenAPI target](#api-schema-for-openapi-and-smithy-target)\n\n**Smithy Model Target** : Smithy is a language for defining services and software development kits (SDKs). Smithy models provide\na more structured approach to defining APIs compared to OpenAPI, and are particularly useful for connecting to AWS services.\nAgentCore Gateway supports built-in AWS service models only. It connects to services using Smithy model definitions\n\n- Supports OAUTH and API_KEY credential providers\n- Ideal for AWS service integrations\n- Need API schema. The construct provide 3 ways to upload a API schema to Smity target\n- When using the default IAM authentication (no `credentialProviderConfigurations` specified), The construct only\n  grants permission to read the Smithy schema file from S3. You MUST manually grant permissions for the gateway\n  role to invoke the actual Smithy API endpoints\n\n> Note: For Smithy model targets that access AWS services, your Gateway's execution role needs permissions to access those services.\nFor example, for a DynamoDB target, your execution role needs permissions to perform DynamoDB operations.\nThis is not managed by the construct due to the large number of options. Please refer to\n[Smithy Model Permission](https://docs.aws.amazon.com/bedrock-agentcore/latest/devguide/gateway-prerequisites-permissions.html) for example.\n\n**MCP Server Target**: Model Context Protocol (MCP) servers provide external tools, data access, and custom functions for AI agents.\nMCP servers enable agents to interact with external systems and services through a standardized protocol. Gateway automatically\ndiscovers and indexes available tools from MCP servers through synchronization.\n\n**Key Features:**\n\n- Requires explicit authentication configuration (OAuth2 recommended, empty array for NoAuth)\n- Ideal for connecting to external MCP-compliant servers\n- The endpoint must use HTTPS protocol\n- Supported MCP protocol versions: 2025-06-18, 2025-03-26\n- Automatic tool discovery through synchronization\n\n**Synchronization Behavior:**\n\nMCP Server targets require synchronization to discover and index available tools:\n\n- **Implicit Synchronization (Automatic)**: Tool discovery happens automatically during:\n  - Target creation (`CreateGatewayTarget`)\n  - Target updates (`UpdateGatewayTarget`)\n  - The Gateway calls the MCP server's `tools/list` endpoint and indexes tools without user intervention\n\n- **Explicit Synchronization (Manual)**: When the MCP server's tools change independently (new tools added, schemas modified, tools removed):\n  - The Gateway's tool catalog becomes stale\n  - Call the `SynchronizeGatewayTargets` API to refresh the catalog\n  - Use the `grantSync()` method to grant permissions to Lambda functions, CI/CD pipelines, or scheduled tasks that will trigger synchronization\n\n**Authentication & Permissions:**\n\nWhen using OAuth2, the Gateway service role automatically receives:\n\n- `bedrock-agentcore:GetWorkloadAccessToken`\n- `bedrock-agentcore:GetResourceOauth2Token`\n- `secretsmanager:GetSecretValue`\n- KMS decrypt (if secrets are encrypted)\n\nFor explicit synchronization, use `grantSync()` to grant `bedrock-agentcore:SynchronizeGatewayTargets` permission to your operator roles.\n\n> For more information, refer to the [MCP Server Target documentation](https://docs.aws.amazon.com/bedrock-agentcore/latest/devguide/gateway-target-MCPservers.html).\n\n### Understanding Tool Naming\n\nWhen tools are exposed through gateway targets, AgentCore Gateway prefixes each tool name with the target name to ensure uniqueness across multiple targets. This is important to understand when building your application logic.\n\n**Naming Pattern:**\n\n**Example:**\n\nIf your target is named `my-lambda-target` and provides a tool called `calculate_price`, agents will discover and invoke it as `my-lambda-target__calculate_price`.\n\n**Important Considerations:**\n\n- **For Lambda Targets**: Your Lambda handler must strip the target name prefix before processing the tool request. The full tool name (with prefix) is sent in the event.\n- **For MCP Server Targets**: The MCP server receives tool calls with the prefixed name from the gateway.\n- **For OpenAPI/Smithy Targets**: The gateway handles the prefix automatically when mapping to API operations based on the `operationId`.\n\nThis naming convention ensures that:\n\n- Tools from different targets don't collide even if they have the same name\n- Agents can access tools from multiple targets through a single gateway\n- Tool names remain unique in the unified tool catalog\n\nFor more details, see the [Gateway Tool Naming Documentation](https://docs.aws.amazon.com/bedrock-agentcore/latest/devguide/gateway-tool-naming.html).\n\n### Tools schema For Lambda target\n\nThe lambda target need tools schema to understand the fuunction lambda provides. You can upload the tool schema by following 3 ways:\n\n- From a local asset file\n\n```typescript\nconst toolSchema = agentcore.ToolSchema.fromLocalAsset(\n    path.join(__dirname, \"schemas\", \"my-tool-schema.json\")\n  );\n```\n\n- From an existing S3 file:\n\n```typescript\n\nconst toolSchema = agentcore.ToolSchema.fromS3File(\n    s3.Bucket.fromBucketName(this, \"SchemasBucket\", \"my-schemas-bucket\"),\n    \"tools/complex-tool-schema.json\",\n    \"123456789012\"\n  );\n```\n\n- From Inline:\n\n```typescript\nconst toolSchema = agentcore.ToolSchema.fromInline([{\n      name: \"hello_world\",\n      description: \"A simple hello world tool\",\n      inputSchema: {\n        type: agentcore.SchemaDefinitionType.OBJECT,\n        properties: {\n          name: {\n            type: agentcore.SchemaDefinitionType.STRING,\n            description: \"The name to greet\",\n          },\n        },\n        required: [\"name\"],\n      },\n    }]);\n\n```\n\n### Api schema For OpenAPI and Smithy target\n\nThe OpenAPI and Smithy target need API Schema. The Gateway construct provide three ways to upload API schema for your target:\n\n- From a local asset file (requires binding to scope):\n\n```typescript fixture=default\n// When using ApiSchema.fromLocalAsset, you must bind the schema to a scope\nconst schema = agentcore.ApiSchema.fromLocalAsset(path.join(__dirname, \"mySchema.yml\"));\n\nschema.bind(this);\n```\n\n- From an inline schema:\n\n```typescript fixture=default\nconst inlineSchema = agentcore.ApiSchema.fromInline(`\nopenapi: 3.0.3\ninfo:\n  title: Library API\n  version: 1.0.0\npaths:\n  /search:\n    get:\n      summary: Search for books\n      operationId: searchBooks\n      parameters:\n        - name: query\n          in: query\n          required: true\n          schema:\n            type: string\n`);\n```\n\n- From an existing S3 file:\n\n```typescript fixture=default\nconst bucket = s3.Bucket.fromBucketName(this, \"ExistingBucket\", \"my-schema-bucket\");\nconst s3Schema = agentcore.ApiSchema.fromS3File(bucket, \"schemas/action-group.yaml\");\n```\n\n### Outbound auth\n\nOutbound authorization lets Amazon Bedrock AgentCore gateways securely access gateway targets on behalf of users authenticated\nand authorized during Inbound Auth.\n\nAgentCore Gateway supports the following types of outbound authorization:\n\n**IAM-based outbound authorization** – The gateway uses its execution role to authenticate with AWS services. This is the default\n and most common approach for Lambda targets and AWS service integrations.\n\n**2-legged OAuth (OAuth 2LO)** – Use OAuth 2.0 two-legged flow (2LO) for targets that require OAuth authentication.\nThe gateway authenticates on its own behalf, not on behalf of a user.\n\n**API key** – Use the AgentCore service/AWS console to generate an API key to authenticate access to the gateway target.\n\n**Note > You need to set up the outbound identity before you can create a gateway target.\n\n### Basic Gateway Target Creation\n\nYou can create targets in two ways: using the static factory methods on `GatewayTarget` or using the convenient `addTarget` methods on the gateway instance.\n\n#### Using addTarget methods (Recommended)\n\nBelow are the examples on how you can create Lambda , Smity and OpenAPI target using `addTarget` method.\n\n```typescript fixture=default\n// Create a gateway first\nconst gateway = new agentcore.Gateway(this, \"MyGateway\", {\n  gatewayName: \"my-gateway\",\n});\n\nconst lambdaFunction = new lambda.Function(this, \"MyFunction\", {\n  runtime: lambda.Runtime.NODEJS_22_X,\n  handler: \"index.handler\",\n  code: lambda.Code.fromInline(`\n    exports.handler = async (event) => {\n      return {\n        statusCode: 200,\n        body: JSON.stringify({ message: 'Hello from Lambda!' })\n      };\n    };\n  `),\n});\n\nconst lambdaTarget = gateway.addLambdaTarget(\"MyLambdaTarget\", {\n  gatewayTargetName: \"my-lambda-target\",\n  description: \"Lambda function target\",\n  lambdaFunction: lambdaFunction,\n  toolSchema: agentcore.ToolSchema.fromInline([\n    {\n      name: \"hello_world\",\n      description: \"A simple hello world tool\",\n      inputSchema: {\n        type: agentcore.SchemaDefinitionType.OBJECT,\n        properties: {\n          name: {\n            type: agentcore.SchemaDefinitionType.STRING,\n            description: \"The name to greet\",\n          },\n        },\n        required: [\"name\"],\n      },\n    },\n  ]),\n});\n```\n\n- OpenAPI Target\n\n``` typescript\nconst gateway = new agentcore.Gateway(this, \"MyGateway\", {\n  gatewayName: \"my-gateway\",\n});\n\n// These ARNs are returned when creating the API key credential provider via Console or API\nconst apiKeyProviderArn = \"arn:aws:bedrock-agentcore:us-east-1:123456789012:token-vault/abc123/apikeycredentialprovider/my-apikey\"\nconst apiKeySecretArn = \"arn:aws:secretsmanager:us-east-1:123456789012:secret:my-apikey-secret-abc123\"\n\nconst bucket = s3.Bucket.fromBucketName(this, \"ExistingBucket\", \"my-schema-bucket\");\nconst s3mySchema = agentcore.ApiSchema.fromS3File(bucket, \"schemas/myschema.yaml\");\n\n// Add an OpenAPI target directly to the gateway\nconst target = gateway.addOpenApiTarget(\"MyTarget\", {\n  gatewayTargetName: \"my-api-target\",\n  description: \"Target for external API integration\",\n  apiSchema: s3mySchema,\n  credentialProviderConfigurations: [\n    agentcore.GatewayCredentialProvider.fromApiKeyIdentityArn({\n      providerArn: apiKeyProviderArn,\n      secretArn: apiKeySecretArn,\n      credentialLocation: agentcore.ApiKeyCredentialLocation.header({\n        credentialParameterName: \"X-API-Key\",\n      }),\n    }),\n  ],\n});\n\n// This make sure your s3 bucket is available before target\ntarget.node.addDependency(bucket);\n```\n\n- Smithy Target\n\n```typescript fixture=default\n\nconst gateway = new agentcore.Gateway(this, \"MyGateway\", {\n  gatewayName: \"my-gateway\",\n});\n\nconst smithySchema = agentcore.ApiSchema.fromLocalAsset(\n  path.join(__dirname, \"models\", \"smithy-model.json\")\n);\nsmithySchema.bind(this);\n\nconst smithyTarget = gateway.addSmithyTarget(\"MySmithyTarget\", {\n  gatewayTargetName: \"my-smithy-target\",\n  description: \"Smithy model target\",\n  smithyModel: smithySchema,\n\n});\n```\n\n- MCP Server Target\n\n```typescript fixture=default\nconst gateway = new agentcore.Gateway(this, \"MyGateway\", {\n  gatewayName: \"my-gateway\",\n});\n\n// OAuth2 authentication (recommended)\n// Note: Create the OAuth provider using AWS console or Identity L2 construct when available\nconst oauthProviderArn = \"arn:aws:bedrock-agentcore:us-east-1:123456789012:token-vault/abc123/oauth2credentialprovider/my-oauth\";\nconst oauthSecretArn = \"arn:aws:secretsmanager:us-east-1:123456789012:secret:my-oauth-secret-abc123\";\n\n// Add an MCP server target directly to the gateway\nconst mcpTarget = gateway.addMcpServerTarget(\"MyMcpServer\", {\n  gatewayTargetName: \"my-mcp-server\",\n  description: \"External MCP server integration\",\n  endpoint: \"https://my-mcp-server.example.com\",\n  credentialProviderConfigurations: [\n    agentcore.GatewayCredentialProvider.fromOauthIdentityArn({\n      providerArn: oauthProviderArn,\n      secretArn: oauthSecretArn,\n      scopes:['mcp-runtime-server/invoke']\n    }),\n  ],\n});\n\n// Grant sync permission to a Lambda function that will trigger synchronization\nconst syncFunction = new lambda.Function(this, \"SyncFunction\", {\n  runtime: lambda.Runtime.PYTHON_3_12,\n  handler: \"index.handler\",\n  code: lambda.Code.fromInline(`\nimport boto3\n\ndef handler(event, context):\n    client = boto3.client('bedrock-agentcore')\n    response = client.synchronize_gateway_targets(\n        gatewayIdentifier=event['gatewayId'],\n        targetIds=[event['targetId']]\n    )\n    return response\n  `),\n});\n\nmcpTarget.grantSync(syncFunction);\n```\n\n#### Using static factory methods\n\nCreate Gateway target using static convienence method.\n\n- Lambda Target\n\n```typescript fixture=default\nconst gateway = new agentcore.Gateway(this, \"MyGateway\", {\n  gatewayName: \"my-gateway\",\n});\n\nconst lambdaFunction = new lambda.Function(this, \"MyFunction\", {\n  runtime: lambda.Runtime.NODEJS_22_X,\n  handler: \"index.handler\",\n  code: lambda.Code.fromInline(`\n        exports.handler = async (event) => {\n            return {\n                statusCode: 200,\n                body: JSON.stringify({ message: 'Hello from Lambda!' })\n            };\n        };\n    `),\n});\n\n// Create a gateway target with Lambda and tool schema\nconst target = agentcore.GatewayTarget.forLambda(this, \"MyLambdaTarget\", {\n  gatewayTargetName: \"my-lambda-target\",\n  description: \"Target for Lambda function integration\",\n  gateway: gateway,\n  lambdaFunction: lambdaFunction,\n  toolSchema: agentcore.ToolSchema.fromLocalAsset(\n    path.join(__dirname, \"schemas\", \"my-tool-schema.json\")\n  ),\n});\n```\n\n- OpenAPI Target\n\n```typescript fixture=default\n\nconst gateway = new agentcore.Gateway(this, \"MyGateway\", {\n  gatewayName: \"my-gateway\",\n});\n\n// outbound auth (Use AWS console to create it, Once Identity L2 construct is available you can use it to create identity)\nconst apiKeyIdentityArn = \"arn:aws:bedrock-agentcore:us-east-1:123456789012:token-vault/abc123/apikeycredentialprovider/my-apikey\"\nconst apiKeySecretArn = \"arn:aws:secretsmanager:us-east-1:123456789012:secret:my-apikey-secret-abc123\"\n\nconst opneapiSchema = agentcore.ApiSchema.fromLocalAsset(path.join(__dirname, \"mySchema.yml\"));\nopneapiSchema.bind(this);\n\n// Create a gateway target with OpenAPI Schema\nconst target = agentcore.GatewayTarget.forOpenApi(this, \"MyTarget\", {\n  gatewayTargetName: \"my-api-target\",\n  description: \"Target for external API integration\",\n  gateway: gateway,  // Note: you need to pass the gateway reference\n  apiSchema: opneapiSchema,\n  credentialProviderConfigurations: [\n    agentcore.GatewayCredentialProvider.fromApiKeyIdentityArn({\n     providerArn: apiKeyIdentityArn,\n     secretArn: apiKeySecretArn\n    }),\n  ],\n});\n\n```\n\n- Smithy Target\n\n```typescript fixture=default\n\nconst gateway = new agentcore.Gateway(this, \"MyGateway\", {\n  gatewayName: \"my-gateway\",\n});\n\nconst smithySchema = agentcore.ApiSchema.fromLocalAsset(\n  path.join(__dirname, \"models\", \"smithy-model.json\")\n);\nsmithySchema.bind(this);\n\n// Create a gateway target with Smithy Model and OAuth\nconst target = agentcore.GatewayTarget.forSmithy(this, \"MySmithyTarget\", {\n  gatewayTargetName: \"my-smithy-target\",\n  description: \"Target for Smithy model integration\",\n  gateway: gateway,\n  smithyModel: smithySchema,\n});\n\n```\n\n- MCP Server Target\n\n```typescript fixture=default\nconst gateway = new agentcore.Gateway(this, \"MyGateway\", {\n  gatewayName: \"my-gateway\",\n});\n\n// OAuth2 authentication (recommended)\n// Note: Create the OAuth provider using AWS console or Identity L2 construct when available\nconst oauthProviderArn = \"arn:aws:bedrock-agentcore:us-east-1:123456789012:token-vault/abc123/oauth2credentialprovider/my-oauth\";\nconst oauthSecretArn = \"arn:aws:secretsmanager:us-east-1:123456789012:secret:my-oauth-secret-abc123\";\n\n// Create a gateway target with MCP Server\nconst mcpTarget = agentcore.GatewayTarget.forMcpServer(this, \"MyMcpServer\", {\n  gatewayTargetName: \"my-mcp-server\",\n  description: \"External MCP server integration\",\n  gateway: gateway,\n  endpoint: \"https://my-mcp-server.example.com\",\n  credentialProviderConfigurations: [\n    agentcore.GatewayCredentialProvider.fromOauthIdentityArn({\n      providerArn: oauthProviderArn,\n      secretArn: oauthSecretArn,\n      scopes:['mcp-runtime-server/invoke']\n    }),\n  ],\n});\n```\n\n### Advanced Usage: Direct Configuration for gateway target\n\nFor advanced use cases where you need full control over the target configuration, you can create configurations manually using the static factory methods and use the GatewayTarget constructor directly.\n\n#### Configuration Factory Methods\n\nEach target type has a corresponding configuration class with a static `create()` method:\n\n- **Lambda**: `LambdaTargetConfiguration.create(lambdaFunction, toolSchema)`\n- **OpenAPI**: `OpenApiTargetConfiguration.create(apiSchema, validateSchema?)`\n- **Smithy**: `SmithyTargetConfiguration.create(smithyModel)`\n\n#### Example: Lambda Target with Custom Configuration\n\n```typescript\nconst gateway = new agentcore.Gateway(this, \"MyGateway\", {\n  gatewayName: \"my-gateway\",\n});\n\nconst myLambdaFunction = new lambda.Function(this, \"MyFunction\", {\n  runtime: lambda.Runtime.NODEJS_22_X,\n  handler: \"index.handler\",\n  code: lambda.Code.fromInline(`\n    exports.handler = async (event) => ({ statusCode: 200 });\n  `),\n});\n\nconst myToolSchema = agentcore.ToolSchema.fromInline([{\n  name: \"my_tool\",\n  description: \"My custom tool\",\n  inputSchema: {\n    type: agentcore.SchemaDefinitionType.OBJECT,\n    properties: {},\n  },\n}]);\n\n// Create a custom Lambda configuration\nconst customConfig = agentcore.LambdaTargetConfiguration.create(\n  myLambdaFunction,\n  myToolSchema\n);\n\n// Use the GatewayTarget constructor directly\nconst target = new agentcore.GatewayTarget(this, \"AdvancedTarget\", {\n  gateway: gateway,\n  gatewayTargetName: \"advanced-target\",\n  targetConfiguration: customConfig,  // Manually created configuration\n  credentialProviderConfigurations: [\n    agentcore.GatewayCredentialProvider.fromIamRole()\n  ]\n});\n```\n\nThis approach gives you full control over the configuration but is typically not necessary for most use cases. The convenience methods (`GatewayTarget.forLambda()`, `GatewayTarget.forOpenApi()`, `GatewayTarget.forSmithy()`) handle all of this internally.\n\n### Gateway Target IAM Permissions\n\nThe Gateway Target construct provides convenient methods for granting IAM permissions:\n\n```typescript fixture=default\n// Create a gateway and target\nconst gateway = new agentcore.Gateway(this, \"MyGateway\", {\n  gatewayName: \"my-gateway\",\n});\n\nconst smithySchema = agentcore.ApiSchema.fromLocalAsset(\n  path.join(__dirname, \"models\", \"smithy-model.json\")\n);\nsmithySchema.bind(this);\n\n// Create a gateway target with Smithy Model and OAuth\nconst target = agentcore.GatewayTarget.forSmithy(this, \"MySmithyTarget\", {\n  gatewayTargetName: \"my-smithy-target\",\n  description: \"Target for Smithy model integration\",\n  gateway: gateway,\n  smithyModel: smithySchema,\n});\n\n// Create a role that needs access to the gateway target\nconst userRole = new iam.Role(this, \"UserRole\", {\n  assumedBy: new iam.ServicePrincipal(\"lambda.amazonaws.com\"),\n});\n\n// Grant read permissions (Get and List actions)\ntarget.grantRead(userRole);\n\n// Grant manage permissions (Create, Update, Delete actions)\ntarget.grantManage(userRole);\n\n// Grant specific custom permissions\ntarget.grant(userRole, \"bedrock-agentcore:GetGatewayTarget\");\n\n\n// Grants permission to invoke this Gateway\ngateway.grantInvoke(userRole);\n```\n\n## Memory\n\nMemory is a critical component of intelligence. While Large Language Models (LLMs) have impressive capabilities, they lack persistent memory across conversations. Amazon Bedrock AgentCore Memory addresses this limitation by providing a managed service that enables AI agents to maintain context over time, remember important facts, and deliver consistent, personalized experiences.\n\nAgentCore Memory operates on two levels:\n\n- **Short-Term Memory**: Immediate conversation context and session-based information that provides continuity within a single interaction or closely related sessions.\n- **Long-Term Memory**: Persistent information extracted and stored across multiple conversations, including facts, preferences, and summaries that enable personalized experiences over time.\n\nWhen you interact with the memory via the `CreateEvent` API, you store interactions in Short-Term Memory (STM) instantly. These interactions can include everything from user messages, assistant responses, to tool actions.\n\nTo write to long-term memory, you need to configure extraction strategies which define how and where to store information from conversations for future use. These strategies are asynchronously processed from raw events after every few turns based on the strategy that was selected. You can't create long term memory records directly, as they are extracted asynchronously by AgentCore Memory.\n\n### Memory Properties\n\n| Name | Type | Required | Description |\n|------|------|----------|-------------|\n| `memoryName` | `string` | Yes | The name of the memory |\n| `expirationDuration` | `Duration` | No | Short-term memory expiration in days (between 7 and 365). Default: 90 days |\n| `description` | `string` | No | Optional description for the memory. Default: no description. |\n| `kmsKey` | `IKey` | No | Custom KMS key to use for encryption. Default: Your data is encrypted with a key that AWS owns and manages for you |\n| `memoryStrategies` | `MemoryStrategyBase[]` | No | Built-in extraction strategies to use for this memory. Default: No extraction strategies (short term memory only) |\n| `executionRole` | `iam.IRole` | No | The IAM role that provides permissions for the memory to access AWS services. Default: A new role will be created. |\n| `tags` | `{ [key: string]: string }` | No | Tags for memory. Default: no tags. |\n\n### Basic Memory Creation\n\nBelow you can find how to configure a simple short-term memory (STM) with no long-term memory extraction strategies.\nNote how you set `expirationDuration`, which defines the time the events will be stored in the short-term memory before they expire.\n\n```typescript fixture=default\n\n// Create a basic memory with default settings, no LTM strategies\nconst memory = new agentcore.Memory(this, \"MyMemory\", {\n  memoryName: \"my_memory\",\n  description: \"A memory for storing user interactions for a period of 90 days\",\n  expirationDuration: cdk.Duration.days(90),\n});\n```\n\nBasic Memory with Custom KMS Encryption\n\n```typescript fixture=default\n// Create a custom KMS key for encryption\nconst encryptionKey = new kms.Key(this, \"MemoryEncryptionKey\", {\n  enableKeyRotation: true,\n  description: \"KMS key for memory encryption\",\n});\n\n// Create memory with custom encryption\nconst memory = new agentcore.Memory(this, \"MyMemory\", {\n  memoryName: \"my_encrypted_memory\",\n  description: \"Memory with custom KMS encryption\",\n  expirationDuration: cdk.Duration.days(90),\n  kmsKey: encryptionKey,\n});\n```\n\n### LTM Memory Extraction Stategies\n\nIf you need long-term memory for context recall across sessions, you can setup memory extraction strategies\nto extract the relevant memory from the raw events.\n\nAmazon Bedrock AgentCore Memory has different memory strategies for extracting and organizing information:\n\n- **Summarization**: to summarize interactions to preserve critical context and key insights.\n- **Semantic Memory**: to extract general factual knowledge, concepts and meanings from raw conversations using vector embeddings.\nThis enables similarity-based retrieval of relevant facts and context.\n- **User Preferences**: to extract user behavior patterns from raw conversations.\n\nYou can use built-in extraction strategies for quick setup, or create custom extraction strategies with specific models and prompt templates.\n\n### Memory with Built-in Strategies\n\nThe library provides three built-in LTM strategies. These are default strategies for organizing and extracting memory data,\neach optimized for specific use cases.\n\nFor example: An agent helps multiple users with cloud storage setup. From these conversations,\nsee how each strategy processes users expressing confusion about account connection:\n\n1. **Summarization Strategy** (`MemoryStrategy.usingBuiltInSummarization()`)\nThis strategy compresses conversations into concise overviews, preserving essential context and key insights for quick recall.\nExtracted memory example: Users confused by cloud setup during onboarding.\n\n   - Extracts concise summaries to preserve critical context and key insights\n   - Namespace: `/strategies/{memoryStrategyId}/actors/{actorId}/sessions/{sessionId}`\n\n2. **Semantic Memory Strategy** (`MemoryStrategy.usingBuiltInSemantic()`)\nDistills general facts, concepts, and underlying meanings from raw conversational data, presenting the information in a context-independent format.\nExtracted memory example: In-context learning = task-solving via examples, no training needed.\n\n   - Extracts general factual knowledge, concepts and meanings from raw conversations\n   - Namespace: `/strategies/{memoryStrategyId}/actors/{actorId}`\n\n3. **User Preference Strategy** (`MemoryStrategy.usingBuiltInUserPreference()`)\nCaptures individual preferences, interaction patterns, and personalized settings to enhance future experiences.\nExtracted memory example: User needs clear guidance on cloud storage account connection during onboarding.\n\n   - Extracts user behavior patterns from raw conversations\n   - Namespace: `/strategies/{memoryStrategyId}/actors/{actorId}`\n\n```typescript fixture=default\n// Create memory with built-in strategies\nconst memory = new agentcore.Memory(this, \"MyMemory\", {\n  memoryName: \"my_memory\",\n  description: \"Memory with built-in strategies\",\n  expirationDuration: cdk.Duration.days(90),\n  memoryStrategies: [\n    agentcore.MemoryStrategy.usingBuiltInSummarization(),\n    agentcore.MemoryStrategy.usingBuiltInSemantic(),\n    agentcore.MemoryStrategy.usingBuiltInUserPreference(),\n  ],\n});\n```\n\nThe name generated for each built in memory strategy is as follows:\n\n- For Summarization: `summary_builtin_cdk001`\n- For Semantic:`semantic_builtin_cdk001>`\n- For User Preferences: `preference_builtin_cdk001`\n\n### Memory with custom Strategies\n\nWith Long-Term Memory, organization is managed through Namespaces.\n\nAn `actor` refers to entity such as end users or agent/user combinations. For example, in a coding support chatbot,\nthe actor is usually the developer asking questions. Using the actor ID helps the system know which user the memory belongs to,\nkeeping each user's data separate and organized.\n\nA `session` is usually a single conversation or interaction period between the user and the AI agent.\nIt groups all related messages and events that happen during that conversation.\n\nA `namespace` is used to logically group and organize long-term memories. It ensures data stays neat, separate, and secure.\n\nWith AgentCore Memory, you need to add a namespace when you define a memory strategy. This namespace helps define where the long-term memory\nwill be logically grouped. Every time a new long-term memory is extracted using this memory strategy, it is saved under the namespace you set.\nThis means that all long-term memories are scoped to their specific namespace, keeping them organized and preventing any mix-ups with other\nusers or sessions. You should use a hierarchical format separated by forward slashes /. This helps keep memories organized clearly. As needed,\nyou can choose to use the below pre-defined variables within braces in the namespace based on your applications' organization needs:\n\n- `actorId` – Identifies who the long-term memory belongs to, such as a user\n- `memoryStrategyId` – Shows which memory strategy is being used. This strategy identifier is auto-generated when you create a memory using CreateMemory operation.\n- `sessionId` – Identifies which session or conversation the memory is from.\n\nFor example, if you define the following namespace as the input to your strategy in CreateMemory operation:\n\n```shell\n/strategy/{memoryStrategyId}/actor/{actorId}/session/{sessionId}\n```\n\nAfter memory creation, this namespace might look like:\n\n```shell\n/strategy/summarization-93483043//actor/actor-9830m2w3/session/session-9330sds8\n```\n\nYou can customise the namespace, i.e. where the memories are stored by using the following methods:\n\n1. **Summarization Strategy** (`MemoryStrategy.usingSummarization(props)`)\n1. **Semantic Memory Strategy** (`MemoryStrategy.usingSemantic(props)`)\n1. **User Preference Strategy** (`MemoryStrategy.usingUserPreference(props)`)\n\n```typescript fixture=default\n// Create memory with built-in strategies\nconst memory = new agentcore.Memory(this, \"MyMemory\", {\n  memoryName: \"my_memory\",\n  description: \"Memory with built-in strategies\",\n  expirationDuration: cdk.Duration.days(90),\n  memoryStrategies: [\n    agentcore.MemoryStrategy.usingUserPreference({\n        name: \"CustomerPreferences\",\n        namespaces: [\"support/customer/{actorId}/preferences\"]\n    }),\n    agentcore.MemoryStrategy.usingSemantic({\n        name: \"CustomerSupportSemantic\",\n        namespaces: [\"support/customer/{actorId}/semantic\"]\n    }),\n  ],\n});\n```\n\nCustom memory strategies let you tailor memory extraction and consolidation to your specific domain or use case.\nYou can override the prompts for extracting and consolidating semantic, summary, or user preferences.\nYou can also choose the model that you want to use for extraction and consolidation.\n\nThe custom prompts you create are appended to a non-editable system prompt.\n\nSince a custom strategy requires you to invoke certain FMs, you need a role with appropriate permissions. For that, you can:\n\n- Let the L2 construct create a minimum permission role for you when use L2 Bedrock Foundation Models.\n- Use a custom role with the overly permissive `AmazonBedrockAgentCoreMemoryBedrockModelInferenceExecutionRolePolicy` managed policy.\n- Use a custom role with your own custom policies.\n\n#### Memory with Custom Execution Role\n\nKeep in mind that memories that **do not** use custom strategies do not require a service role.\nSo even if you provide it, it will be ignored as it will never be used.\n\n```typescript fixture=default\n// Create a custom execution role\nconst executionRole = new iam.Role(this, \"MemoryExecutionRole\", {\n  assumedBy: new iam.ServicePrincipal(\"bedrock-agentcore.amazonaws.com\"),\n  managedPolicies: [\n    iam.ManagedPolicy.fromAwsManagedPolicyName(\n      \"AmazonBedrockAgentCoreMemoryBedrockModelInferenceExecutionRolePolicy\"\n    ),\n  ],\n});\n\n// Create memory with custom execution role\nconst memory = new agentcore.Memory(this, \"MyMemory\", {\n  memoryName: \"my_memory\",\n  description: \"Memory with custom execution role\",\n  expirationDuration: cdk.Duration.days(90),\n  executionRole: executionRole,\n});\n```\n\nIn customConsolidation and customExtraction, the model property uses the [@aws-cdk/aws-bedrock-alph](https://www.npmjs.com/package/@aws-cdk/aws-bedrock-alpha) library which must be installed separately.\n\n```typescript fixture=default\n// Create a custom semantic memory strategy\nconst customSemanticStrategy = agentcore.MemoryStrategy.usingSemantic({\n  name: \"customSemanticStrategy\",\n  description: \"Custom semantic memory strategy\",\n  namespaces: [\"/custom/strategies/{memoryStrategyId}/actors/{actorId}\"],\n  customConsolidation: {\n    model: bedrock.BedrockFoundationModel.ANTHROPIC_CLAUDE_3_5_SONNET_V1_0,\n    appendToPrompt: \"Custom consolidation prompt for semantic memory\",\n  },\n  customExtraction: {\n    model: bedrock.BedrockFoundationModel.ANTHROPIC_CLAUDE_3_5_SONNET_V1_0,\n    appendToPrompt: \"Custom extraction prompt for semantic memory\",\n  },\n});\n\n// Create memory with custom strategy\nconst memory = new agentcore.Memory(this, \"MyMemory\", {\n  memoryName: \"my-custom-memory\",\n  description: \"Memory with custom strategy\",\n  expirationDuration: cdk.Duration.days(90),\n  memoryStrategies: [customSemanticStrategy],\n});\n```\n\n### Memory with self-managed Strategies\n\nA self-managed strategy in Amazon Bedrock AgentCore Memory gives you complete control over your memory extraction and consolidation pipelines.\nWith a self-managed strategy, you can build custom memory processing workflows while leveraging Amazon Bedrock AgentCore for storage and retrieval.\n\nFor additional information, you can refer to the [developer guide for self managed strategies](https://docs.aws.amazon.com/bedrock-agentcore/latest/devguide/memory-self-managed-strategies.html).\n\nCreate the required AWS resources including:\n\n- an S3 bucket in your account where Amazon Bedrock AgentCore will deliver batched event payloads.\n- an SNS topic for job notifications. Use FIFO topics if processing order within sessions is important for your use case.\n\nThe construct will apply the correct permissions to the memory execution role to access these resources.\n\n```typescript fixture=default\n\nconst bucket = new s3.Bucket(this, 'memoryBucket', {\n  bucketName: 'test-memory',\n  removalPolicy: cdk.RemovalPolicy.DESTROY,\n  autoDeleteObjects: true,\n});\n\nconst topic = new sns.Topic(this, 'topic');\n\n// Create a custom semantic memory strategy\nconst selfManagedStrategy = agentcore.MemoryStrategy.usingSelfManaged({\n  name: \"selfManagedStrategy\",\n  description: \"self managed memory strategy\",\n  historicalContextWindowSize: 5,\n  invocationConfiguration: {\n    topic: topic,\n    s3Location: {\n      bucketName: bucket.bucketName,\n      objectKey: 'memory/',\n    }\n  },\n  triggerConditions: {\n    messageBasedTrigger: 1,\n    timeBasedTrigger: cdk.Duration.seconds(10),\n    tokenBasedTrigger: 100\n  }\n});\n\n// Create memory with custom strategy\nconst memory = new agentcore.Memory(this, \"MyMemory\", {\n  memoryName: \"my-custom-memory\",\n  description: \"Memory with custom strategy\",\n  expirationDuration: cdk.Duration.days(90),\n  memoryStrategies: [selfManagedStrategy],\n});\n```\n\n### Memory Strategy Methods\n\nYou can add new memory strategies to the memory construct using the `addMemoryStrategy()` method, for instance:\n\n```typescript fixture=default\n// Create memory without initial strategies\nconst memory = new agentcore.Memory(this, \"test-memory\", {\n  memoryName: \"test_memory_add_strategy\",\n  description: \"A test memory for testing addMemoryStrategy method\",\n  expirationDuration: cdk.Duration.days(90),\n});\n\n// Add strategies after instantiation\nmemory.addMemoryStrategy(agentcore.MemoryStrategy.usingBuiltInSummarization());\nmemory.addMemoryStrategy(agentcore.MemoryStrategy.usingBuiltInSemantic());\n```\n"
+    "markdown": "# Amazon Bedrock AgentCore Construct Library\n\n<!--BEGIN STABILITY BANNER-->\n\n---\n\n![cdk-constructs: Experimental](https://img.shields.io/badge/cdk--constructs-experimental-important.svg?style=for-the-badge)\n\n> The APIs of higher level constructs in this module are experimental and under active development.\n> They are subject to non-backward compatible changes or removal in any future version. These are\n> not subject to the [Semantic Versioning](https://semver.org/) model and breaking changes will be\n> announced in the release notes. This means that while you may use them, you may need to update\n> your source code when upgrading to a newer version of this package.\n\n---\n\n<!--END STABILITY BANNER-->\n\n| **Language**                                                                                   | **Package**                             |\n| :--------------------------------------------------------------------------------------------- | --------------------------------------- |\n| ![Typescript Logo](https://docs.aws.amazon.com/cdk/api/latest/img/typescript32.png) TypeScript | `@aws-cdk/aws-bedrock-agentcore-alpha` |\n\n[Amazon Bedrock AgentCore](https://aws.amazon.com/bedrock/agentcore/) enables you to deploy and operate highly capable AI agents securely, at scale. It offers infrastructure purpose-built for dynamic agent workloads, powerful tools to enhance agents, and essential controls for real-world deployment. AgentCore services can be used together or independently and work with any framework including CrewAI, LangGraph, LlamaIndex, and Strands Agents, as well as any foundation model in or outside of Amazon Bedrock, giving you ultimate flexibility. AgentCore eliminates the undifferentiated heavy lifting of building specialized agent infrastructure, so you can accelerate agents to production.\n\nThis construct library facilitates the deployment of Bedrock AgentCore primitives, enabling you to create sophisticated AI applications that can interact with your systems and data sources.\n\n> **Note:** Users need to ensure their CDK deployment role has the `iam:CreateServiceLinkedRole` permission for AgentCore service-linked roles.\n\n## Table of contents\n\n- [AgentCore Runtime](#agentcore-runtime)\n  - [Runtime Versioning](#runtime-versioning)\n  - [Runtime Endpoints](#runtime-endpoints)\n  - [AgentCore Runtime Properties](#agentcore-runtime-properties)\n  - [Runtime Endpoint Properties](#runtime-endpoint-properties)\n  - [Creating a Runtime](#creating-a-runtime)\n    - [Option 1: Use an existing image in ECR](#option-1-use-an-existing-image-in-ecr)\n    - [Managing Endpoints and Versions](#managing-endpoints-and-versions)\n    - [Option 2: Use a local asset](#option-2-use-a-local-asset)\n- [Browser Custom tool](#browser)\n  - [Browser properties](#browser-properties)\n  - [Browser Network modes](#browser-network-modes)\n  - [Basic Browser Creation](#basic-browser-creation)\n  - [Browser IAM permissions](#browser-iam-permissions)\n- [Code Interpreter Custom tool](#code-interpreter)\n  - [Code Interpreter properties](#code-interpreter-properties)\n  - [Code Interpreter Network Modes](#code-interpreter-network-modes)\n  - [Basic Code Interpreter Creation](#basic-code-interpreter-creation)\n  - [Code Interpreter IAM permissions](#code-interpreter-iam-permissions)\n- [Gateway](#gateway)\n  - [Gateway Properties](#gateway-properties)\n  - [Basic Gateway Creation](#basic-gateway-creation)\n  - [Protocol configuration](#protocol-configuration)\n  - [Inbound authorization](#inbound-authorization)\n  - [Gateway with KMS Encryption](#gateway-with-kms-encryption)\n  - [Gateway with Custom Execution Role](#gateway-with-custom-execution-role)\n  - [Gateway IAM Permissions](#gateway-iam-permissions)\n- [Gateway Target](#gateway-target)\n  - [Gateway Target Properties](#gateway-target-properties)\n  - [Targets types](#targets-types)\n  - [Outbound auth](#outbound-auth)\n  - [Api schema](#api-schema-for-openapi-and-smithy-target)\n  - [Basic Gateway Target Creation](#basic-gateway-target-creation)\n    - [Using addTarget methods (Recommended)](#using-addtarget-methods-recommended)\n    - [Using static factory methods](#using-static-factory-methods)\n  - [Lambda Target with Tool Schema](#tools-schema-for-lambda-target)\n  - [Smithy Model Target with OAuth](#api-schema-for-openapi-and-smithy-target)\n  - [Gateway Target IAM Permissions](#gateway-target-iam-permissions)\n- [Memory](#memory)\n  - [Memory properties](#memory-properties)\n  - [Basic Memory Creation](#basic-memory-creation)\n  - [LTM Memory Extraction Stategies](#ltm-memory-extraction-stategies)\n  - [Memory Strategy Methods](#memory-strategy-methods)\n- [Amazon Bedrock AgentCore Construct Library](#amazon-bedrock-agentcore-construct-library)\n  - [Table of contents](#table-of-contents)\n  - [AgentCore Runtime](#agentcore-runtime)\n    - [Runtime Endpoints](#runtime-endpoints)\n    - [AgentCore Runtime Properties](#agentcore-runtime-properties)\n    - [Runtime Endpoint Properties](#runtime-endpoint-properties)\n    - [Creating a Runtime](#creating-a-runtime)\n      - [Option 1: Use an existing image in ECR](#option-1-use-an-existing-image-in-ecr)\n      - [Option 2: Use a local asset](#option-2-use-a-local-asset)\n      - [Option 3: Use direct code deployment](#option-3-use-direct-code-deployment)\n    - [Granting Permissions to Invoke Bedrock Models or Inference Profiles](#granting-permissions-to-invoke-bedrock-models-or-inference-profiles)\n    - [Runtime Versioning](#runtime-versioning)\n      - [Managing Endpoints and Versions](#managing-endpoints-and-versions)\n        - [Step 1: Initial Deployment](#step-1-initial-deployment)\n        - [Step 2: Creating Custom Endpoints](#step-2-creating-custom-endpoints)\n        - [Step 3: Runtime Update Deployment](#step-3-runtime-update-deployment)\n        - [Step 4: Testing with Staging Endpoints](#step-4-testing-with-staging-endpoints)\n        - [Step 5: Promoting to Production](#step-5-promoting-to-production)\n    - [Creating Standalone Runtime Endpoints](#creating-standalone-runtime-endpoints)\n      - [Example: Creating an endpoint for an existing runtime](#example-creating-an-endpoint-for-an-existing-runtime)\n    - [Runtime Authentication Configuration](#runtime-authentication-configuration)\n      - [IAM Authentication (Default)](#iam-authentication-default)\n      - [Cognito Authentication](#cognito-authentication)\n      - [JWT Authentication](#jwt-authentication)\n      - [OAuth Authentication](#oauth-authentication)\n      - [Using a Custom IAM Role](#using-a-custom-iam-role)\n    - [Runtime Network Configuration](#runtime-network-configuration)\n      - [Public Network Mode (Default)](#public-network-mode-default)\n      - [VPC Network Mode](#vpc-network-mode)\n      - [Managing Security Groups with VPC Configuration](#managing-security-groups-with-vpc-configuration)\n  - [Browser](#browser)\n    - [Browser Network modes](#browser-network-modes)\n    - [Browser Properties](#browser-properties)\n    - [Basic Browser Creation](#basic-browser-creation)\n    - [Browser with Tags](#browser-with-tags)\n    - [Browser with VPC](#browser-with-vpc)\n    - [Browser with Recording Configuration](#browser-with-recording-configuration)\n    - [Browser with Custom Execution Role](#browser-with-custom-execution-role)\n    - [Browser with S3 Recording and Permissions](#browser-with-s3-recording-and-permissions)\n    - [Browser IAM Permissions](#browser-iam-permissions)\n  - [Code Interpreter](#code-interpreter)\n    - [Code Interpreter Network Modes](#code-interpreter-network-modes)\n    - [Code Interpreter Properties](#code-interpreter-properties)\n    - [Basic Code Interpreter Creation](#basic-code-interpreter-creation)\n    - [Code Interpreter with VPC](#code-interpreter-with-vpc)\n    - [Code Interpreter with Sandbox Network Mode](#code-interpreter-with-sandbox-network-mode)\n    - [Code Interpreter with Custom Execution Role](#code-interpreter-with-custom-execution-role)\n    - [Code Interpreter IAM Permissions](#code-interpreter-iam-permissions)\n    - [Code interpreter with tags](#code-interpreter-with-tags)\n  - [Memory](#memory)\n    - [Memory Properties](#memory-properties)\n    - [Basic Memory Creation](#basic-memory-creation)\n    - [LTM Memory Extraction Stategies](#ltm-memory-extraction-stategies)\n    - [Memory with Built-in Strategies](#memory-with-built-in-strategies)\n    - [Memory with custom Strategies](#memory-with-custom-strategies)\n      - [Memory with Custom Execution Role](#memory-with-custom-execution-role)\n    - [Memory with self-managed Strategies](#memory-with-self-managed-strategies)\n    - [Memory Strategy Methods](#memory-strategy-methods)\n\n## AgentCore Runtime\n\nThe AgentCore Runtime construct enables you to deploy containerized agents on Amazon Bedrock AgentCore.\nThis L2 construct simplifies runtime creation just pass your ECR repository name\nand the construct handles all the configuration with sensible defaults.\n\n### Runtime Endpoints\n\nEndpoints provide a stable way to invoke specific versions of your agent runtime, enabling controlled deployments across different environments.\nWhen you create an agent runtime, Amazon Bedrock AgentCore automatically creates a \"DEFAULT\" endpoint which always points to the latest version\nof runtime.\n\nYou can create additional endpoints in two ways:\n\n1. **Using Runtime.addEndpoint()** - Convenient method when creating endpoints alongside the runtime.\n2. **Using RuntimeEndpoint** - Flexible approach for existing runtimes.\n\nFor example, you might keep a \"production\" endpoint on a stable version while testing newer versions\nthrough a \"staging\" endpoint. This separation allows you to test changes thoroughly before promoting them\nto production by simply updating the endpoint to point to the newer version.\n\n### AgentCore Runtime Properties\n\n| Name | Type | Required | Description |\n|------|------|----------|-------------|\n| `runtimeName` | `string` | Yes | The name of the agent runtime. Valid characters are a-z, A-Z, 0-9, _ (underscore). Must start with a letter and can be up to 48 characters long |\n| `agentRuntimeArtifact` | `AgentRuntimeArtifact` | Yes | The artifact configuration for the agent runtime containing the container configuration with ECR URI |\n| `executionRole` | `iam.IRole` | No | The IAM role that provides permissions for the agent runtime. If not provided, a role will be created automatically |\n| `networkConfiguration` | `NetworkConfiguration` | No | Network configuration for the agent runtime. Defaults to `RuntimeNetworkConfiguration.usingPublicNetwork()` |\n| `description` | `string` | No | Optional description for the agent runtime |\n| `protocolConfiguration` | `ProtocolType` | No | Protocol configuration for the agent runtime. Defaults to `ProtocolType.HTTP` |\n| `authorizerConfiguration` | `RuntimeAuthorizerConfiguration` | No | Authorizer configuration for the agent runtime. Use `RuntimeAuthorizerConfiguration` static methods to create configurations for IAM, Cognito, JWT, or OAuth authentication |\n| `environmentVariables` | `{ [key: string]: string }` | No | Environment variables for the agent runtime. Maximum 50 environment variables |\n| `tags` | `{ [key: string]: string }` | No | Tags for the agent runtime. A list of key:value pairs of tags to apply to this Runtime resource |\n| `lifecycleConfiguration` | LifecycleConfiguration | No | The life cycle configuration for the AgentCore Runtime. Defaults to 900 seconds (15 minutes) for idle, 28800 seconds (8 hours) for max life time |\n| `requestHeaderConfiguration` | RequestHeaderConfiguration | No | Configuration for HTTP request headers that will be passed through to the runtime. Defaults to no configuration |\n\n### Runtime Endpoint Properties\n\n| Name | Type | Required | Description |\n|------|------|----------|-------------|\n| `endpointName` | `string` | Yes | The name of the runtime endpoint. Valid characters are a-z, A-Z, 0-9, _ (underscore). Must start with a letter and can be up to 48 characters long |\n| `agentRuntimeId` | `string` | Yes | The Agent Runtime ID for this endpoint |\n| `agentRuntimeVersion` | `string` | Yes | The Agent Runtime version for this endpoint. Must be between 1 and 5 characters long.|\n| `description` | `string` | No | Optional description for the runtime endpoint |\n| `tags` | `{ [key: string]: string }` | No | Tags for the runtime endpoint |\n\n### Creating a Runtime\n\n#### Option 1: Use an existing image in ECR\n\nReference an image available within ECR.\n\n```typescript fixture=default\nconst repository = new ecr.Repository(this, \"TestRepository\", {\n  repositoryName: \"test-agent-runtime\",\n});\n\n// The runtime by default create ECR permission only for the repository available in the account the stack is being deployed\nconst agentRuntimeArtifact = agentcore.AgentRuntimeArtifact.fromEcrRepository(repository, \"v1.0.0\");\n\n// Create runtime using the built image\nconst runtime = new agentcore.Runtime(this, \"MyAgentRuntime\", {\n  runtimeName: \"myAgent\",\n  agentRuntimeArtifact: agentRuntimeArtifact\n});\n```\n\n#### Option 2: Use a local asset\n\nReference a local directory containing a Dockerfile.\nImages are built from a local Docker context directory (with a Dockerfile), uploaded to Amazon Elastic Container Registry (ECR)\nby the CDK toolkit,and can be naturally referenced in your CDK app.\n\n```typescript fixture=default\nconst agentRuntimeArtifact = agentcore.AgentRuntimeArtifact.fromAsset(\n  path.join(__dirname, \"path to agent dockerfile directory\")\n);\n\nconst runtime = new agentcore.Runtime(this, \"MyAgentRuntime\", {\n  runtimeName: \"myAgent\",\n  agentRuntimeArtifact: agentRuntimeArtifact,\n});\n```\n\n#### Option 3: Use direct code deployment\n\nWith the container deployment method, developers create a Dockerfile, build ARM-compatible containers, manage ECR repositories, and upload containers for code changes. This works well where container DevOps pipelines have already been established to automate deployments.\n\nHowever, customers looking for fully managed deployments can benefit from direct code deployment, which can significantly improve developer time and productivity. Direct code deployment provides a secure and scalable path forward for rapid prototyping agent capabilities to deploying production workloads at scale.\n\nWith direct code deployment, developers create a zip archive of code and dependencies, upload to Amazon S3, and configure the bucket in the agent configuration. A ZIP archive containing Linux arm64 dependencies needs to be uploaded to S3 as a pre-requisite to Create Agent Runtime.\n\nFor more information, please refer to the [documentation](https://docs.aws.amazon.com/bedrock-agentcore/latest/devguide/runtime-get-started-code-deploy.html).\n\n```typescript fixture=default\n// S3 bucket containing the agent core\nconst codeBucket = new s3.Bucket(this, \"AgentCode\", {\n  bucketName: \"my-code-bucket\",\n  removalPolicy: RemovalPolicy.DESTROY, // For demo purposes\n});\n\n// the bucket above needs to contain the agent code\n\nconst agentRuntimeArtifact = agentcore.AgentRuntimeArtifact.fromS3(\n  {\n    bucketName: codeBucket.bucketName,\n    objectKey: 'deployment_package.zip',\n  },\n  agentcore.AgentCoreRuntime.PYTHON_3_12,\n  ['opentelemetry-instrument', 'main.py']\n);\n\nconst runtimeInstance = new agentcore.Runtime(this, \"MyAgentRuntime\", {\n  runtimeName: \"myAgent\",\n  agentRuntimeArtifact: agentRuntimeArtifact,\n});\n```\n\n#### Option 4: Use an ECR container image URI\n\nReference an ECR container image directly by its URI. This is useful when you have a pre-existing ECR image URI from CloudFormation parameters or cross-stack references. No IAM permissions are automatically granted - you must ensure the runtime has ECR pull permissions.\n\n```typescript fixture=default\n// Direct URI reference\nconst agentRuntimeArtifact = agentcore.AgentRuntimeArtifact.fromImageUri(\n  \"123456789012.dkr.ecr.us-east-1.amazonaws.com/my-agent:v1.0.0\"\n);\n\nconst runtime = new agentcore.Runtime(this, \"MyAgentRuntime\", {\n  runtimeName: \"myAgent\",\n  agentRuntimeArtifact: agentRuntimeArtifact,\n});\n```\n\nYou can also use CloudFormation parameters or references:\n\n```typescript fixture=default\n// Using a CloudFormation parameter\nconst imageUriParam = new cdk.CfnParameter(this, \"ImageUri\", {\n  type: \"String\",\n  description: \"Container image URI for the agent runtime\",\n});\n\nconst agentRuntimeArtifact = agentcore.AgentRuntimeArtifact.fromImageUri(\n  imageUriParam.valueAsString\n);\n\nconst runtime = new agentcore.Runtime(this, \"MyAgentRuntime\", {\n  runtimeName: \"myAgent\",\n  agentRuntimeArtifact: agentRuntimeArtifact,\n});\n```\n\n### Granting Permissions to Invoke Bedrock Models or Inference Profiles\n\nTo grant the runtime permissions to invoke Bedrock models or inference profiles:\n\n```typescript fixture=default\n// Note: This example uses @aws-cdk/aws-bedrock-alpha which must be installed separately\ndeclare const runtime: agentcore.Runtime;\n\n// Define the Bedrock Foundation Model\nconst model = bedrock.BedrockFoundationModel.ANTHROPIC_CLAUDE_3_7_SONNET_V1_0;\n\n// Grant the runtime permissions to invoke the model\nmodel.grantInvoke(runtime);\n\n// Create a cross-region inference profile for Claude 3.7 Sonnet\nconst inferenceProfile = bedrock.CrossRegionInferenceProfile.fromConfig({\n  geoRegion: bedrock.CrossRegionInferenceProfileRegion.US,\n  model: bedrock.BedrockFoundationModel.ANTHROPIC_CLAUDE_3_7_SONNET_V1_0\n});\n\n// Grant the runtime permissions to invoke the inference profile\ninferenceProfile.grantInvoke(runtime);\n```\n\n### Runtime Versioning\n\nAmazon Bedrock AgentCore automatically manages runtime versioning to ensure safe deployments and rollback capabilities.\nWhen you create an agent runtime, AgentCore automatically creates version 1 (V1). Each subsequent update to the\nruntime configuration (such as updating the container image, modifying network settings, or changing protocol configurations)\ncreates a new immutable version. These versions contain complete, self-contained configurations that can be referenced by endpoints,\nallowing you to maintain different versions for different environments or gradually roll out updates.\n\n#### Managing Endpoints and Versions\n\nAmazon Bedrock AgentCore automatically manages runtime versioning to provide safe deployments and rollback capabilities. You can follow\nthe steps below to understand how to use versioning with runtime for controlled deployments across different environments.\n\n##### Step 1: Initial Deployment\n\nWhen you first create an agent runtime, AgentCore automatically creates Version 1 of your runtime. At this point, a DEFAULT endpoint is\nautomatically created that points to Version 1. This DEFAULT endpoint serves as the main access point for your runtime.\n\n```typescript fixture=default\nconst repository = new ecr.Repository(this, \"TestRepository\", {\n  repositoryName: \"test-agent-runtime\",\n});\n\nconst runtime = new agentcore.Runtime(this, \"MyAgentRuntime\", {\n  runtimeName: \"myAgent\",\n  agentRuntimeArtifact: agentcore.AgentRuntimeArtifact.fromEcrRepository(repository, \"v1.0.0\"),\n});\n```\n\n##### Step 2: Creating Custom Endpoints\n\nAfter the initial deployment, you can create additional endpoints for different environments. For example, you might create a \"production\"\nendpoint that explicitly points to Version 1. This allows you to maintain stable access points for specific environments while keeping the\nflexibility to test newer versions elsewhere.\n\n```typescript fixture=default\nconst repository = new ecr.Repository(this, \"TestRepository\", {\n  repositoryName: \"test-agent-runtime\",\n});\n\nconst runtime = new agentcore.Runtime(this, \"MyAgentRuntime\", {\n  runtimeName: \"myAgent\",\n  agentRuntimeArtifact: agentcore.AgentRuntimeArtifact.fromEcrRepository(repository, \"v1.0.0\"),\n});\n\nconst prodEndpoint = runtime.addEndpoint(\"production\", {\n  version: \"1\",\n  description: \"Stable production endpoint - pinned to v1\"\n});\n```\n\n##### Step 3: Runtime Update Deployment\n\nWhen you update the runtime configuration (such as updating the container image, modifying network settings, or changing protocol\nconfigurations), AgentCore automatically creates a new version (Version 2). Upon this update:\n\n- Version 2 is created automatically with the new configuration\n- The DEFAULT endpoint automatically updates to point to Version 2\n- Any explicitly pinned endpoints (like the production endpoint) remain on their specified versions\n\n```typescript fixture=default\nconst repository = new ecr.Repository(this, \"TestRepository\", {\n  repositoryName: \"test-agent-runtime\",\n});\n\nconst agentRuntimeArtifactNew = agentcore.AgentRuntimeArtifact.fromEcrRepository(repository, \"v2.0.0\");\n\nconst runtime = new agentcore.Runtime(this, \"MyAgentRuntime\", {\n  runtimeName: \"myAgent\",\n  agentRuntimeArtifact: agentRuntimeArtifactNew,\n});\n```\n\n##### Step 4: Testing with Staging Endpoints\n\nOnce Version 2 exists, you can create a staging endpoint that points to the new version. This staging endpoint allows you to test the\nnew version in a controlled environment before promoting it to production. This separation ensures that production traffic continues\nto use the stable version while you validate the new version.\n\n```typescript fixture=default\nconst repository = new ecr.Repository(this, \"TestRepository\", {\n  repositoryName: \"test-agent-runtime\",\n});\n\nconst agentRuntimeArtifactNew = agentcore.AgentRuntimeArtifact.fromEcrRepository(repository, \"v2.0.0\");\n\nconst runtime = new agentcore.Runtime(this, \"MyAgentRuntime\", {\n  runtimeName: \"myAgent\",\n  agentRuntimeArtifact: agentRuntimeArtifactNew,\n});\n\nconst stagingEndpoint = runtime.addEndpoint(\"staging\", {\n  version: \"2\",\n  description: \"Staging environment for testing new version\"\n});\n```\n\n##### Step 5: Promoting to Production\n\nAfter thoroughly testing the new version through the staging endpoint, you can update the production endpoint to point to Version 2.\nThis controlled promotion process ensures that you can validate changes before they affect production traffic.\n\n```typescript fixture=default\nconst repository = new ecr.Repository(this, \"TestRepository\", {\n  repositoryName: \"test-agent-runtime\",\n});\n\nconst agentRuntimeArtifactNew = agentcore.AgentRuntimeArtifact.fromEcrRepository(repository, \"v2.0.0\");\n\nconst runtime = new agentcore.Runtime(this, \"MyAgentRuntime\", {\n  runtimeName: \"myAgent\",\n  agentRuntimeArtifact: agentRuntimeArtifactNew,\n});\n\nconst prodEndpoint = runtime.addEndpoint(\"production\", {\n  version: \"2\",  // New version added here\n  description: \"Stable production endpoint\"\n});\n```\n\n### Creating Standalone Runtime Endpoints\n\nRuntimeEndpoint can also be created as a standalone resource.\n\n#### Example: Creating an endpoint for an existing runtime\n\n```typescript fixture=default\n// Reference an existing runtime by its ID\nconst existingRuntimeId = \"abc123-runtime-id\"; // The ID of an existing runtime\n\n// Create a standalone endpoint\nconst endpoint = new agentcore.RuntimeEndpoint(this, \"MyEndpoint\", {\n  endpointName: \"production\",\n  agentRuntimeId: existingRuntimeId,\n  agentRuntimeVersion: \"1\", // Specify which version to use\n  description: \"Production endpoint for existing runtime\"\n});\n```\n\n### Runtime Authentication Configuration\n\nThe AgentCore Runtime supports multiple authentication modes to secure access to your agent endpoints. Authentication is configured during runtime creation using the `RuntimeAuthorizerConfiguration` class's static factory methods.\n\n#### IAM Authentication (Default)\n\nIAM authentication is the default mode, when no authorizerConfiguration is set then the underlying service use IAM.\n\n#### Cognito Authentication\n\nTo configure AWS Cognito User Pool authentication:\n\n```typescript fixture=default\ndeclare const userPool: cognito.UserPool;\ndeclare const userPoolClient: cognito.UserPoolClient;\ndeclare const anotherUserPoolClient: cognito.UserPoolClient;\n\nconst repository = new ecr.Repository(this, \"TestRepository\", {\n  repositoryName: \"test-agent-runtime\",\n});\nconst agentRuntimeArtifact = agentcore.AgentRuntimeArtifact.fromEcrRepository(repository, \"v1.0.0\");\n\nconst runtime = new agentcore.Runtime(this, \"MyAgentRuntime\", {\n  runtimeName: \"myAgent\",\n  agentRuntimeArtifact: agentRuntimeArtifact,\n  authorizerConfiguration: agentcore.RuntimeAuthorizerConfiguration.usingCognito(\n    userPool, // User Pool (required)\n    [userPoolClient, anotherUserPoolClient], // User Pool Clients\n  ),\n});\n```\n\n#### JWT Authentication\n\nTo configure custom JWT authentication with your own OpenID Connect (OIDC) provider:\n\n```typescript fixture=default\nconst repository = new ecr.Repository(this, \"TestRepository\", {\n  repositoryName: \"test-agent-runtime\",\n});\nconst agentRuntimeArtifact = agentcore.AgentRuntimeArtifact.fromEcrRepository(repository, \"v1.0.0\");\n\nconst runtime = new agentcore.Runtime(this, \"MyAgentRuntime\", {\n  runtimeName: \"myAgent\",\n  agentRuntimeArtifact: agentRuntimeArtifact,\n  authorizerConfiguration: agentcore.RuntimeAuthorizerConfiguration.usingJWT(\n    \"https://example.com/.well-known/openid-configuration\",  // Discovery URL (required)\n    [\"client1\", \"client2\"],  // Allowed Client IDs (optional)\n    [\"audience1\"]           // Allowed Audiences (optional)\n  ),\n});\n```\n\n**Note**: The discovery URL must end with `/.well-known/openid-configuration`.\n\n#### OAuth Authentication\n\nTo configure OAuth 2.0 authentication:\n\n```typescript fixture=default\nconst repository = new ecr.Repository(this, \"TestRepository\", {\n  repositoryName: \"test-agent-runtime\",\n});\nconst agentRuntimeArtifact = agentcore.AgentRuntimeArtifact.fromEcrRepository(repository, \"v1.0.0\");\n\nconst runtime = new agentcore.Runtime(this, \"MyAgentRuntime\", {\n  runtimeName: \"myAgent\",\n  agentRuntimeArtifact: agentRuntimeArtifact,\n  authorizerConfiguration: agentcore.RuntimeAuthorizerConfiguration.usingOAuth(\n    \"https://github.com/.well-known/openid-configuration\",\n    \"oauth_client_123\",\n  ),\n});\n```\n\n#### Using a Custom IAM Role\n\nInstead of using the auto-created execution role, you can provide your own IAM role with specific permissions:\nThe auto-created role includes all necessary baseline permissions for ECR access, CloudWatch logging, and X-Ray tracing. When providing a custom role, ensure these permissions are included.\n\n### Runtime Network Configuration\n\nThe AgentCore Runtime supports two network modes for deployment:\n\n#### Public Network Mode (Default)\n\nBy default, runtimes are deployed in PUBLIC network mode, which provides internet access suitable for less sensitive or open-use scenarios:\n\n```typescript fixture=default\nconst repository = new ecr.Repository(this, \"TestRepository\", {\n  repositoryName: \"test-agent-runtime\",\n});\nconst agentRuntimeArtifact = agentcore.AgentRuntimeArtifact.fromEcrRepository(repository, \"v1.0.0\");\n\n// Explicitly using public network (this is the default)\nconst runtime = new agentcore.Runtime(this, \"MyAgentRuntime\", {\n  runtimeName: \"myAgent\",\n  agentRuntimeArtifact: agentRuntimeArtifact,\n  networkConfiguration: agentcore.RuntimeNetworkConfiguration.usingPublicNetwork(),\n});\n```\n\n#### VPC Network Mode\n\nFor enhanced security and network isolation, you can deploy your runtime within a VPC:\n\n```typescript fixture=default\nconst repository = new ecr.Repository(this, \"TestRepository\", {\n  repositoryName: \"test-agent-runtime\",\n});\nconst agentRuntimeArtifact = agentcore.AgentRuntimeArtifact.fromEcrRepository(repository, \"v1.0.0\");\n\n// Create or use an existing VPC\nconst vpc = new ec2.Vpc(this, 'MyVpc', {\n  maxAzs: 2,\n});\n\n// Configure runtime with VPC\nconst runtime = new agentcore.Runtime(this, \"MyAgentRuntime\", {\n  runtimeName: \"myAgent\",\n  agentRuntimeArtifact: agentRuntimeArtifact,\n  networkConfiguration: agentcore.RuntimeNetworkConfiguration.usingVpc(this, {\n    vpc: vpc,\n    vpcSubnets: { subnetType: ec2.SubnetType.PRIVATE_WITH_EGRESS },\n    // Optionally specify security groups, or one will be created automatically\n    // securityGroups: [mySecurityGroup],\n  }),\n});\n\n```\n\n#### Managing Security Groups with VPC Configuration\n\nWhen using VPC mode, the Runtime implements `ec2.IConnectable`, allowing you to manage network access using the `connections` property:\n\n```typescript fixture=default\nconst vpc = new ec2.Vpc(this, 'MyVpc', {\n  maxAzs: 2,\n});\n\nconst repository = new ecr.Repository(this, \"TestRepository\", {\n  repositoryName: \"test-agent-runtime\",\n});\nconst agentRuntimeArtifact = agentcore.AgentRuntimeArtifact.fromEcrRepository(repository, \"v1.0.0\");\n\n// Create runtime with VPC configuration\nconst runtime = new agentcore.Runtime(this, \"MyAgentRuntime\", {\n  runtimeName: \"myAgent\",\n  agentRuntimeArtifact: agentRuntimeArtifact,\n  networkConfiguration: agentcore.RuntimeNetworkConfiguration.usingVpc(this, {\n    vpc: vpc,\n    vpcSubnets: { subnetType: ec2.SubnetType.PRIVATE_WITH_EGRESS },\n  }),\n});\n\n// Now you can manage network access using the connections property\n// Allow inbound HTTPS traffic from a specific security group\nconst webServerSecurityGroup = new ec2.SecurityGroup(this, 'WebServerSG', { vpc });\nruntime.connections.allowFrom(webServerSecurityGroup, ec2.Port.tcp(443), 'Allow HTTPS from web servers');\n\n// Allow outbound connections to a database\nconst databaseSecurityGroup = new ec2.SecurityGroup(this, 'DatabaseSG', { vpc });\nruntime.connections.allowTo(databaseSecurityGroup, ec2.Port.tcp(5432), 'Allow PostgreSQL connection');\n\n// Allow outbound HTTPS to anywhere (for external API calls)\nruntime.connections.allowToAnyIpv4(ec2.Port.tcp(443), 'Allow HTTPS outbound');\n```\n\n### Runtime IAM Permissions\n\nThe Runtime construct provides convenient methods for granting IAM permissions to principals that need to invoke the runtime or manage its execution role.\n\n```typescript fixture=default\nconst repository = new ecr.Repository(this, \"TestRepository\", {\n  repositoryName: \"test-agent-runtime\",\n});\nconst agentRuntimeArtifact = agentcore.AgentRuntimeArtifact.fromEcrRepository(repository, \"v1.0.0\");\n\n// Create a runtime\nconst runtime = new agentcore.Runtime(this, \"MyRuntime\", {\n  runtimeName: \"my_runtime\",\n  agentRuntimeArtifact: agentRuntimeArtifact,\n});\n\n// Create a Lambda function that needs to invoke the runtime\nconst invokerFunction = new lambda.Function(this, \"InvokerFunction\", {\n  runtime: lambda.Runtime.PYTHON_3_12,\n  handler: \"index.handler\",\n  code: lambda.Code.fromInline(`\nimport boto3\ndef handler(event, context):\n    client = boto3.client('bedrock-agentcore')\n    # Invoke the runtime...\n  `),\n});\n\n// Grant permission to invoke the runtime directly\nruntime.grantInvokeRuntime(invokerFunction);\n\n// Grant permission to invoke the runtime on behalf of a user\n// (requires X-Amzn-Bedrock-AgentCore-Runtime-User-Id header)\nruntime.grantInvokeRuntimeForUser(invokerFunction);\n\n// Grant both invoke permissions (most common use case)\nruntime.grantInvoke(invokerFunction);\n\n// Grant specific custom permissions to the runtime's execution role\nruntime.grant(['bedrock:InvokeModel'], ['arn:aws:bedrock:*:*:*']);\n\n// Add a policy statement to the runtime's execution role\nruntime.addToRolePolicy(new iam.PolicyStatement({\n  actions: ['s3:GetObject'],\n  resources: ['arn:aws:s3:::my-bucket/*'],\n}));\n```\n\n### Other configuration\n\n#### Lifecycle configuration\n\nThe LifecycleConfiguration input parameter to CreateAgentRuntime lets you manage the lifecycle of runtime sessions and resources in Amazon Bedrock AgentCore Runtime. This configuration helps optimize resource utilization by automatically cleaning up idle sessions and preventing long-running instances from consuming resources indefinitely.\n\nYou can configure:\n\n- idleRuntimeSessionTimeout: Timeout in seconds for idle runtime sessions. When a session remains idle for this duration, it will trigger termination. Termination can last up to 15 seconds due to logging and other process completion. Default: 900 seconds (15 minutes)\n- maxLifetime: Maximum lifetime for the instance in seconds. Once reached, instances will initialize termination. Termination can last up to 15 seconds due to logging and other process completion. Default: 28800 seconds (8 hours)\n\nFor additional information, please refer to the [documentation](https://docs.aws.amazon.com/bedrock-agentcore/latest/devguide/runtime-lifecycle-settings.html).\n\n```typescript fixture=default\nconst repository = new ecr.Repository(this, \"TestRepository\", {\n  repositoryName: \"test-agent-runtime\",\n});\n\nconst agentRuntimeArtifact = agentcore.AgentRuntimeArtifact.fromEcrRepository(repository, \"v1.0.0\");\n\nnew agentcore.Runtime(this, 'test-runtime', {\n  runtimeName: 'test_runtime',\n  agentRuntimeArtifact: agentRuntimeArtifact,\n  lifecycleConfiguration: {\n    idleRuntimeSessionTimeout: Duration.minutes(10),\n    maxLifetime: Duration.hours(4),\n  },\n});\n```\n\n#### Request header configuration\n\nCustom headers let you pass contextual information from your application directly to your agent code without cluttering the main request payload. This includes authentication tokens like JWT (JSON Web Tokens, which contain user identity and authorization claims) through the Authorization header, allowing your agent to make decisions based on who is calling it. You can also pass custom metadata like user preferences, session identifiers, or trace context using headers prefixed with X-Amzn-Bedrock-AgentCore-Runtime-Custom-, giving your agent access to up to 20 pieces of runtime context that travel alongside each request. This information can be also used in downstream systems like AgentCore Memory that you can namespace based on those characteristics like user_id or aud in claims like line of business.\n\nFor additional information, please refer to the [documentation](https://docs.aws.amazon.com/bedrock-agentcore/latest/devguide/runtime-header-allowlist.html).\n\n```typescript fixture=default\nconst repository = new ecr.Repository(this, \"TestRepository\", {\n  repositoryName: \"test-agent-runtime\",\n});\n\nconst agentRuntimeArtifact = agentcore.AgentRuntimeArtifact.fromEcrRepository(repository, \"v1.0.0\");\n\nnew agentcore.Runtime(this, 'test-runtime', {\n  runtimeName: 'test_runtime',\n  agentRuntimeArtifact: agentRuntimeArtifact,\n  requestHeaderConfiguration: {\n    allowlistedHeaders: ['X-Amzn-Bedrock-AgentCore-Runtime-Custom-H1'],\n  },\n});\n```\n\n## Browser\n\nThe Amazon Bedrock AgentCore Browser provides a secure, cloud-based browser that enables AI agents to interact with websites. It includes security features such as session isolation, built-in observability through live viewing, CloudTrail logging, and session replay capabilities.\n\nAdditional information about the browser tool can be found in the [official documentation](https://docs.aws.amazon.com/bedrock-agentcore/latest/devguide/browser-tool.html)\n\n### Browser Network modes\n\nThe Browser construct supports the following network modes:\n\n1. **Public Network Mode** (`BrowserNetworkMode.usingPublicNetwork()`) - Default\n\n   - Allows internet access for web browsing and external API calls\n   - Suitable for scenarios where agents need to interact with publicly available websites\n   - Enables full web browsing capabilities\n   - VPC mode is not supported with this option\n\n2. **VPC (Virtual Private Cloud)** (`BrowserNetworkMode.usingVpc()`)\n\n   - Select whether to run the browser in a virtual private cloud (VPC).\n   - By configuring VPC connectivity, you enable secure access to private resources such as databases, internal APIs, and services within your VPC.\n\n    While the VPC itself is mandatory, these are optional:\n    - Subnets - if not provided, CDK will select appropriate subnets from the VPC\n    - Security Groups - if not provided, CDK will create a default security group\n    - Specific subnet selection criteria - you can let CDK choose automatically\n\nFor more information on VPC connectivity for Amazon Bedrock AgentCore Browser, please refer to the [official documentation](https://docs.aws.amazon.com/bedrock-agentcore/latest/devguide/agentcore-vpc.html).\n\n### Browser Properties\n\n| Name | Type | Required | Description |\n|------|------|----------|-------------|\n| `browserCustomName` | `string` | Yes | The name of the browser. Must start with a letter and can be up to 48 characters long. Pattern: `[a-zA-Z][a-zA-Z0-9_]{0,47}` |\n| `description` | `string` | No | Optional description for the browser. Can have up to 200 characters |\n| `networkConfiguration` | `BrowserNetworkConfiguration` | No | Network configuration for browser. Defaults to PUBLIC network mode |\n| `recordingConfig` | `RecordingConfig` | No | Recording configuration for browser. Defaults to no recording |\n| `executionRole` | `iam.IRole` | No | The IAM role that provides permissions for the browser to access AWS services. A new role will be created if not provided |\n| `tags` | `{ [key: string]: string }` | No | Tags to apply to the browser resource |\n| `browserSigning` | BrowserSigning | No | Browser signing configuration. Defaults to DISABLED |\n\n### Basic Browser Creation\n\n```typescript fixture=default\n// Create a basic browser with public network access\nconst browser = new agentcore.BrowserCustom(this, \"MyBrowser\", {\n  browserCustomName: \"my_browser\",\n  description: \"A browser for web automation\",\n});\n```\n\n### Browser with Tags\n\n```typescript fixture=default\n// Create a browser with custom tags\nconst browser = new agentcore.BrowserCustom(this, \"MyBrowser\", {\n  browserCustomName: \"my_browser\",\n  description: \"A browser for web automation with tags\",\n  networkConfiguration: agentcore.BrowserNetworkConfiguration.usingPublicNetwork(),\n  tags: {\n    Environment: \"Production\",\n    Team: \"AI/ML\",\n    Project: \"AgentCore\",\n  },\n});\n```\n\n### Browser with VPC\n\n```typescript fixture=default\nconst browser = new agentcore.BrowserCustom(this, 'BrowserVpcWithRecording', {\n  browserCustomName: 'browser_recording',\n  networkConfiguration: agentcore.BrowserNetworkConfiguration.usingVpc(this, {\n    vpc: new ec2.Vpc(this, 'VPC', { restrictDefaultSecurityGroup: false }),\n  }),\n});\n```\n\nBrowser exposes a [connections](https://docs.aws.amazon.com/cdk/api/v2/docs/aws-cdk-lib.aws_ec2.Connections.html) property. This property returns a connections object, which simplifies the process of defining and managing ingress and egress rules for security groups in your AWS CDK applications. Instead of directly manipulating security group rules, you interact with the Connections object of a construct, which then translates your connectivity requirements into the appropriate security group rules. For instance:\n\n```typescript fixture=default\nconst vpc = new ec2.Vpc(this, 'testVPC');\n\nconst browser = new agentcore.BrowserCustom(this, 'test-browser', {\n  browserCustomName: 'test_browser',\n  networkConfiguration: agentcore.BrowserNetworkConfiguration.usingVpc(this, {\n    vpc: vpc,\n  }),\n});\n\nbrowser.connections.addSecurityGroup(new ec2.SecurityGroup(this, 'AdditionalGroup', { vpc }));\n```\n\nSo security groups can be added after the browser construct creation. You can use methods like allowFrom() and allowTo() to grant ingress access to/egress access from a specified peer over a given portRange. The Connections object automatically adds the necessary ingress or egress rules to the security group(s) associated with the calling construct.\n\n### Browser with Recording Configuration\n\n```typescript fixture=default\n// Create an S3 bucket for recordings\nconst recordingBucket = new s3.Bucket(this, \"RecordingBucket\", {\n  bucketName: \"my-browser-recordings\",\n  removalPolicy: RemovalPolicy.DESTROY, // For demo purposes\n});\n\n// Create browser with recording enabled\nconst browser = new agentcore.BrowserCustom(this, \"MyBrowser\", {\n  browserCustomName: \"my_browser\",\n  description: \"Browser with recording enabled\",\n  networkConfiguration: agentcore.BrowserNetworkConfiguration.usingPublicNetwork(),\n  recordingConfig: {\n    enabled: true,\n    s3Location: {\n      bucketName: recordingBucket.bucketName,\n      objectKey: \"browser-recordings/\",\n    },\n  },\n});\n```\n\n### Browser with Custom Execution Role\n\n```typescript fixture=default\n// Create a custom execution role\nconst executionRole = new iam.Role(this, \"BrowserExecutionRole\", {\n  assumedBy: new iam.ServicePrincipal(\"bedrock-agentcore.amazonaws.com\"),\n  managedPolicies: [\n    iam.ManagedPolicy.fromAwsManagedPolicyName(\"AmazonBedrockAgentCoreBrowserExecutionRolePolicy\"),\n  ],\n});\n\n// Create browser with custom execution role\nconst browser = new agentcore.BrowserCustom(this, \"MyBrowser\", {\n  browserCustomName: \"my_browser\",\n  description: \"Browser with custom execution role\",\n  networkConfiguration: agentcore.BrowserNetworkConfiguration.usingPublicNetwork(),\n  executionRole: executionRole,\n});\n```\n\n### Browser with S3 Recording and Permissions\n\n```typescript fixture=default\n// Create an S3 bucket for recordings\nconst recordingBucket = new s3.Bucket(this, \"RecordingBucket\", {\n  bucketName: \"my-browser-recordings\",\n  removalPolicy: RemovalPolicy.DESTROY, // For demo purposes\n});\n\n// Create browser with recording enabled\nconst browser = new agentcore.BrowserCustom(this, \"MyBrowser\", {\n  browserCustomName: \"my_browser\",\n  description: \"Browser with recording enabled\",\n  networkConfiguration: agentcore.BrowserNetworkConfiguration.usingPublicNetwork(),\n  recordingConfig: {\n    enabled: true,\n    s3Location: {\n      bucketName: recordingBucket.bucketName,\n      objectKey: \"browser-recordings/\",\n    },\n  },\n});\n\n// The browser construct automatically grants S3 permissions to the execution role\n// when recording is enabled, so no additional IAM configuration is needed\n```\n\n### Browser with Browser signing\n\nAI agents need to browse the web on your behalf. When your agent visits a website to gather information, complete a form, or verify data, it encounters the same defenses designed to stop unwanted bots: CAPTCHAs, rate limits, and outright blocks.\n\nAmazon Bedrock AgentCore Browser supports Web Bot Auth. Web Bot Auth is a draft IETF protocol that gives agents verifiable cryptographic identities. When you enable Web Bot Auth in AgentCore Browser, the service issues cryptographic credentials that websites can verify. The agent presents these credentials with every request. The WAF may now additionally check the signature, confirm it matches a trusted directory, and allow the request through if verified bots are allowed by the domain owner and other WAF checks are clear.\n\nTo enable the browser to sign requests using the Web Bot Auth protocol, create a browser tool with the browserSigning configuration:\n\n```typescript fixture=default\nconst browser = new agentcore.BrowserCustom(this, 'test-browser', {\n  browserCustomName: 'test_browser',\n  browserSigning: agentcore.BrowserSigning.ENABLED\n});\n```\n\n### Browser IAM Permissions\n\nThe Browser construct provides convenient methods for granting IAM permissions:\n\n```typescript fixture=default\n// Create a browser\nconst browser = new agentcore.BrowserCustom(this, \"MyBrowser\", {\n  browserCustomName: \"my_browser\",\n  description: \"Browser for web automation\",\n  networkConfiguration: agentcore.BrowserNetworkConfiguration.usingPublicNetwork(),\n});\n\n// Create a role that needs access to the browser\nconst userRole = new iam.Role(this, \"UserRole\", {\n  assumedBy: new iam.ServicePrincipal(\"lambda.amazonaws.com\"),\n});\n\n// Grant read permissions (Get and List actions)\nbrowser.grantRead(userRole);\n\n// Grant use permissions (Start, Update, Stop actions)\nbrowser.grantUse(userRole);\n\n// Grant specific custom permissions\nbrowser.grant(userRole, \"bedrock-agentcore:GetBrowserSession\");\n```\n\n## Code Interpreter\n\nThe Amazon Bedrock AgentCore Code Interpreter enables AI agents to write and execute code securely in sandbox environments, enhancing their accuracy and expanding their ability to solve complex end-to-end tasks. This is critical in Agentic AI applications where the agents may execute arbitrary code that can lead to data compromise or security risks. The AgentCore Code Interpreter tool provides secure code execution, which helps you avoid running into these issues.\n\nFor more information about code interpreter, please refer to the [official documentation](https://docs.aws.amazon.com/bedrock-agentcore/latest/devguide/code-interpreter-tool.html)\n\n### Code Interpreter Network Modes\n\nThe Code Interpreter construct supports the following network modes:\n\n1. **Public Network Mode** (`CodeInterpreterNetworkMode.usingPublicNetwork()`) - Default\n\n   - Allows internet access for package installation and external API calls\n   - Suitable for development and testing environments\n   - Enables downloading Python packages from PyPI\n\n2. **Sandbox Network Mode** (`CodeInterpreterNetworkMode.usingSandboxNetwork()`)\n   - Isolated network environment with no internet access\n   - Suitable for production environments with strict security requirements\n   - Only allows access to pre-installed packages and local resources\n\n3. **VPC (Virtual Private Cloud)** (`CodeInterpreterNetworkMode.usingVpc()`)\n   - Select whether to run the browser in a virtual private cloud (VPC).\n   - By configuring VPC connectivity, you enable secure access to private resources such as databases, internal APIs, and services within your VPC.\n\n    While the VPC itself is mandatory, these are optional:\n    - Subnets - if not provided, CDK will select appropriate subnets from the VPC\n    - Security Groups - if not provided, CDK will create a default security group\n    - Specific subnet selection criteria - you can let CDK choose automatically\n\nFor more information on VPC connectivity for Amazon Bedrock AgentCore Browser, please refer to the [official documentation](https://docs.aws.amazon.com/bedrock-agentcore/latest/devguide/agentcore-vpc.html).\n\n### Code Interpreter Properties\n\n| Name | Type | Required | Description |\n|------|------|----------|-------------|\n| `codeInterpreterCustomName` | `string` | Yes | The name of the code interpreter. Must start with a letter and can be up to 48 characters long. Pattern: `[a-zA-Z][a-zA-Z0-9_]{0,47}` |\n| `description` | `string` | No | Optional description for the code interpreter. Can have up to 200 characters |\n| `executionRole` | `iam.IRole` | No | The IAM role that provides permissions for the code interpreter to access AWS services. A new role will be created if not provided |\n| `networkConfiguration` | `CodeInterpreterNetworkConfiguration` | No | Network configuration for code interpreter. Defaults to PUBLIC network mode |\n| `tags` | `{ [key: string]: string }` | No | Tags to apply to the code interpreter resource |\n\n### Basic Code Interpreter Creation\n\n```typescript fixture=default\n// Create a basic code interpreter with public network access\nconst codeInterpreter = new agentcore.CodeInterpreterCustom(this, \"MyCodeInterpreter\", {\n  codeInterpreterCustomName: \"my_code_interpreter\",\n  description: \"A code interpreter for Python execution\",\n});\n```\n\n### Code Interpreter with VPC\n\n```typescript fixture=default\nconst codeInterpreter = new agentcore.CodeInterpreterCustom(this, \"MyCodeInterpreter\", {\n  codeInterpreterCustomName: \"my_sandbox_interpreter\",\n  description: \"Code interpreter with isolated network access\",\n  networkConfiguration: agentcore.BrowserNetworkConfiguration.usingVpc(this, {\n    vpc: new ec2.Vpc(this, 'VPC', { restrictDefaultSecurityGroup: false }),\n  }),\n});\n```\n\nCode Interpreter exposes a [connections](https://docs.aws.amazon.com/cdk/api/v2/docs/aws-cdk-lib.aws_ec2.Connections.html) property. This property returns a connections object, which simplifies the process of defining and managing ingress and egress rules for security groups in your AWS CDK applications. Instead of directly manipulating security group rules, you interact with the Connections object of a construct, which then translates your connectivity requirements into the appropriate security group rules. For instance:\n\n```typescript fixture=default\nconst vpc = new ec2.Vpc(this, 'testVPC');\n\nconst codeInterpreter = new agentcore.CodeInterpreterCustom(this, \"MyCodeInterpreter\", {\n  codeInterpreterCustomName: \"my_sandbox_interpreter\",\n  description: \"Code interpreter with isolated network access\",\n  networkConfiguration: agentcore.BrowserNetworkConfiguration.usingVpc(this, {\n    vpc: vpc,\n  }),\n});\n\ncodeInterpreter.connections.addSecurityGroup(new ec2.SecurityGroup(this, 'AdditionalGroup', { vpc }));\n```\n\nSo security groups can be added after the browser construct creation. You can use methods like allowFrom() and allowTo() to grant ingress access to/egress access from a specified peer over a given portRange. The Connections object automatically adds the necessary ingress or egress rules to the security group(s) associated with the calling construct.\n\n### Code Interpreter with Sandbox Network Mode\n\n```typescript fixture=default\n// Create code interpreter with sandbox network mode (isolated)\nconst codeInterpreter = new agentcore.CodeInterpreterCustom(this, \"MyCodeInterpreter\", {\n  codeInterpreterCustomName: \"my_sandbox_interpreter\",\n  description: \"Code interpreter with isolated network access\",\n  networkConfiguration: agentcore.CodeInterpreterNetworkConfiguration.usingSandboxNetwork(),\n});\n```\n\n### Code Interpreter with Custom Execution Role\n\n```typescript fixture=default\n// Create a custom execution role\nconst executionRole = new iam.Role(this, \"CodeInterpreterExecutionRole\", {\n  assumedBy: new iam.ServicePrincipal(\"bedrock-agentcore.amazonaws.com\"),\n});\n\n// Create code interpreter with custom execution role\nconst codeInterpreter = new agentcore.CodeInterpreterCustom(this, \"MyCodeInterpreter\", {\n  codeInterpreterCustomName: \"my_code_interpreter\",\n  description: \"Code interpreter with custom execution role\",\n  networkConfiguration: agentcore.CodeInterpreterNetworkConfiguration.usingPublicNetwork(),\n  executionRole: executionRole,\n});\n```\n\n### Code Interpreter IAM Permissions\n\nThe Code Interpreter construct provides convenient methods for granting IAM permissions:\n\n```typescript fixture=default\n// Create a code interpreter\nconst codeInterpreter = new agentcore.CodeInterpreterCustom(this, \"MyCodeInterpreter\", {\n  codeInterpreterCustomName: \"my_code_interpreter\",\n  description: \"Code interpreter for Python execution\",\n  networkConfiguration: agentcore.CodeInterpreterNetworkConfiguration.usingPublicNetwork(),\n});\n\n// Create a role that needs access to the code interpreter\nconst userRole = new iam.Role(this, \"UserRole\", {\n  assumedBy: new iam.ServicePrincipal(\"lambda.amazonaws.com\"),\n});\n\n// Grant read permissions (Get and List actions)\ncodeInterpreter.grantRead(userRole);\n\n// Grant use permissions (Start, Invoke, Stop actions)\ncodeInterpreter.grantUse(userRole);\n\n// Grant specific custom permissions\ncodeInterpreter.grant(userRole, \"bedrock-agentcore:GetCodeInterpreterSession\");\n```\n\n### Code interpreter with tags\n\n```typescript fixture=default\n// Create code interpreter with sandbox network mode (isolated)\nconst codeInterpreter = new agentcore.CodeInterpreterCustom(this, \"MyCodeInterpreter\", {\n  codeInterpreterCustomName: \"my_sandbox_interpreter\",\n  description: \"Code interpreter with isolated network access\",\n  networkConfiguration: agentcore.CodeInterpreterNetworkConfiguration.usingPublicNetwork(),\n  tags: {\n    Environment: \"Production\",\n    Team: \"AI/ML\",\n    Project: \"AgentCore\",\n  },\n});\n```\n\n## Gateway\n\nThe Gateway construct provides a way to create Amazon Bedrock Agent Core Gateways, which serve as integration points between agents and external services.\n\n### Gateway Properties\n\n| Name | Type | Required | Description |\n|------|------|----------|-------------|\n| `gatewayName` | `string` | Yes | The name of the gateway. Valid characters are a-z, A-Z, 0-9, _ (underscore) and - (hyphen). Maximum 100 characters |\n| `description` | `string` | No | Optional description for the gateway. Maximum 200 characters |\n| `protocolConfiguration` | `IGatewayProtocolConfig` | No | The protocol configuration for the gateway. Defaults to MCP protocol |\n| `authorizerConfiguration` | `IGatewayAuthorizerConfig` | No | The authorizer configuration for the gateway. Defaults to Cognito |\n| `exceptionLevel` | `GatewayExceptionLevel` | No | The verbosity of exception messages. Use DEBUG mode to see granular exception messages |\n| `kmsKey` | `kms.IKey` | No | The AWS KMS key used to encrypt data associated with the gateway |\n| `role` | `iam.IRole` | No | The IAM role that provides permissions for the gateway to access AWS services. A new role will be created if not provided |\n| `tags` | `{ [key: string]: string }` | No | Tags for the gateway. A list of key:value pairs of tags to apply to this Gateway resource |\n\n### Basic Gateway Creation\n\nThe protocol configuration defaults to MCP and the inbound auth configuration uses Cognito (it is automatically created on your behalf).\n\n```typescript fixture=default\n// Create a basic gateway with default MCP protocol and Cognito authorizer\nconst gateway = new agentcore.Gateway(this, \"MyGateway\", {\n  gatewayName: \"my-gateway\",\n});\n```\n\n### Protocol configuration\n\nCurrently MCP is the only protocol available. To configure it, use the `protocol` property with `McpProtocolConfiguration`:\n\n- Instructions: Guidance for how to use the gateway with your tools\n- Semantic search: Smart tool discovery that finds the right tools without typical limits. It improves accuracy by finding relevant tools based on context\n- Supported versions: Which MCP protocol versions the gateway can use\n\n```typescript fixture=default\nconst gateway = new agentcore.Gateway(this, \"MyGateway\", {\n  gatewayName: \"my-gateway\",\n  protocolConfiguration: new agentcore.McpProtocolConfiguration({\n    instructions: \"Use this gateway to connect to external MCP tools\",\n    searchType: agentcore.McpGatewaySearchType.SEMANTIC,\n    supportedVersions: [agentcore.MCPProtocolVersion.MCP_2025_03_26],\n  }),\n});\n```\n\n### Inbound authorization\n\nBefore you create your gateway, you must set up inbound authorization. Inbound authorization validates users who attempt to access targets through\nyour AgentCore gateway. By default, if not provided, the construct will create and configure Cognito as the default identity provider\n(inbound Auth setup). AgentCore supports the following types of inbound authorization:\n\n**JSON Web Token (JWT)** – A secure and compact token used for authorization. After creating the JWT, you specify it as the authorization\nconfiguration when you create the gateway. You can create a JWT with any of the identity providers at Provider setup and configuration.\n\nYou can configure a custom authorization provider using the `inboundAuthorizer` property with `GatewayAuthorizer.usingCustomJwt()`.\nYou need to specify an OAuth discovery server and client IDs/audiences when you create the gateway. You can specify the following:\n\n- Discovery Url — String that must match the pattern ^.+/\\.well-known/openid-configuration$ for OpenID Connect discovery URLs\n- At least one of the below options depending on the chosen identity provider.\n- Allowed audiences — List of allowed audiences for JWT tokens\n- Allowed clients — List of allowed client identifiers\n\n```typescript fixture=default\nconst gateway = new agentcore.Gateway(this, \"MyGateway\", {\n  gatewayName: \"my-gateway\",\n  authorizerConfiguration: agentcore.GatewayAuthorizer.usingCustomJwt({\n    discoveryUrl: \"https://auth.example.com/.well-known/openid-configuration\",\n    allowedAudience: [\"my-app\"],\n    allowedClients: [\"my-client-id\"],\n  }),\n});\n```\n\n**IAM** – Authorizes through the credentials of the AWS IAM identity trying to access the gateway.\n\n```typescript fixture=default\nconst gateway = new agentcore.Gateway(this, \"MyGateway\", {\n  gatewayName: \"my-gateway\",\n  authorizerConfiguration: agentcore.GatewayAuthorizer.usingAwsIam(),\n});\n\n// Grant access to a Lambda function's role\nconst lambdaRole = new iam.Role(this, \"LambdaRole\", {\n  assumedBy: new iam.ServicePrincipal(\"lambda.amazonaws.com\"),\n});\n\n// The Lambda needs permission to invoke the gateway\ngateway.grantInvoke(lambdaRole);\n```\n\n**Cognito with M2M (Machine-to-Machine) Authentication (Default)** – When no authorizer is specified, the construct automatically creates a Cognito User Pool configured for OAuth 2.0 client credentials flow. This enables machine-to-machine authentication suitable for AI agents and service-to-service communication.\n\nFor more information, see [Setting up Amazon Cognito for Gateway inbound authorization](https://docs.aws.amazon.com/bedrock-agentcore/latest/devguide/identity-idp-cognito.html).\n\n```typescript fixture=default\n// Create a gateway with default Cognito M2M authorizer\nconst gateway = new agentcore.Gateway(this, \"MyGateway\", {\n  gatewayName: \"my-gateway\",\n});\n\n// Access the Cognito resources for authentication setup\nconst userPool = gateway.userPool;\nconst userPoolClient = gateway.userPoolClient;\n\n// Get the token endpoint URL and OAuth scopes for client credentials flow\nconst tokenEndpointUrl = gateway.tokenEndpointUrl;\nconst oauthScopes = gateway.oauthScopes;\n// oauthScopes are in the format: ['{resourceServerId}/read', '{resourceServerId}/write']\n```\n\nTo authenticate with the gateway, request an access token using the client credentials flow and use it to call Gateway endpoints. For more information about the token endpoint, see [The token issuer endpoint](https://docs.aws.amazon.com/cognito/latest/developerguide/token-endpoint.html).\n\nThe following is an example of a token request using curl:\n\n```bash\ncurl -X POST \"${TOKEN_ENDPOINT_URL}\" \\\n  -H \"Content-Type: application/x-www-form-urlencoded\" \\\n  -d \"grant_type=client_credentials\" \\\n  -d \"client_id=${USER_POOL_CLIENT_ID}\" \\\n  -d \"client_secret=${CLIENT_SECRET}\" \\\n  -d \"scope=${OAUTH_SCOPES}\"\n```\n\n### Gateway with KMS Encryption\n\nYou can provide a KMS key, and configure the authorizer as well as the protocol configuration.\n\n```typescript fixture=default\n// Create a KMS key for encryption\nconst encryptionKey = new kms.Key(this, \"GatewayEncryptionKey\", {\n  enableKeyRotation: true,\n  description: \"KMS key for gateway encryption\",\n});\n\n// Create gateway with KMS encryption\nconst gateway = new agentcore.Gateway(this, \"MyGateway\", {\n  gatewayName: \"my-encrypted-gateway\",\n  description: \"Gateway with KMS encryption\",\n  protocolConfiguration: new agentcore.McpProtocolConfiguration({\n    instructions: \"Use this gateway to connect to external MCP tools\",\n    searchType: agentcore.McpGatewaySearchType.SEMANTIC,\n    supportedVersions: [agentcore.MCPProtocolVersion.MCP_2025_03_26],\n  }),\n  authorizerConfiguration: agentcore.GatewayAuthorizer.usingCustomJwt({\n    discoveryUrl: \"https://auth.example.com/.well-known/openid-configuration\",\n    allowedAudience: [\"my-app\"],\n    allowedClients: [\"my-client-id\"],\n  }),\n  kmsKey: encryptionKey,\n  exceptionLevel: agentcore.GatewayExceptionLevel.DEBUG,\n});\n```\n\n### Gateway with Custom Execution Role\n\n```typescript fixture=default\n// Create a custom execution role\nconst executionRole = new iam.Role(this, \"GatewayExecutionRole\", {\n  assumedBy: new iam.ServicePrincipal(\"bedrock-agentcore.amazonaws.com\"),\n  managedPolicies: [\n    iam.ManagedPolicy.fromAwsManagedPolicyName(\"AmazonBedrockAgentCoreGatewayExecutionRolePolicy\"),\n  ],\n});\n\n// Create gateway with custom execution role\nconst gateway = new agentcore.Gateway(this, \"MyGateway\", {\n  gatewayName: \"my-gateway\",\n  description: \"Gateway with custom execution role\",\n  protocolConfiguration: new agentcore.McpProtocolConfiguration({\n    instructions: \"Use this gateway to connect to external MCP tools\",\n    searchType: agentcore.McpGatewaySearchType.SEMANTIC,\n    supportedVersions: [agentcore.MCPProtocolVersion.MCP_2025_03_26],\n  }),\n  authorizerConfiguration: agentcore.GatewayAuthorizer.usingCustomJwt({\n    discoveryUrl: \"https://auth.example.com/.well-known/openid-configuration\",\n    allowedAudience: [\"my-app\"],\n    allowedClients: [\"my-client-id\"],\n  }),\n  role: executionRole,\n});\n```\n\n### Gateway IAM Permissions\n\nThe Gateway construct provides convenient methods for granting IAM permissions:\n\n```typescript fixture=default\n// Create a gateway\nconst gateway = new agentcore.Gateway(this, \"MyGateway\", {\n  gatewayName: \"my-gateway\",\n  description: \"Gateway for external service integration\",\n  protocolConfiguration: new agentcore.McpProtocolConfiguration({\n    instructions: \"Use this gateway to connect to external MCP tools\",\n    searchType: agentcore.McpGatewaySearchType.SEMANTIC,\n    supportedVersions: [agentcore.MCPProtocolVersion.MCP_2025_03_26],\n  }),\n  authorizerConfiguration: agentcore.GatewayAuthorizer.usingCustomJwt({\n    discoveryUrl: \"https://auth.example.com/.well-known/openid-configuration\",\n    allowedAudience: [\"my-app\"],\n    allowedClients: [\"my-client-id\"],\n  }),\n});\n\n// Create a role that needs access to the gateway\nconst userRole = new iam.Role(this, \"UserRole\", {\n  assumedBy: new iam.ServicePrincipal(\"lambda.amazonaws.com\"),\n});\n\n// Grant read permissions (Get and List actions)\ngateway.grantRead(userRole);\n\n// Grant manage permissions (Create, Update, Delete actions)\ngateway.grantManage(userRole);\n\n// Grant specific custom permissions\ngateway.grant(userRole, \"bedrock-agentcore:GetGateway\");\n```\n\n## Gateway Target\n\nAfter Creating gateways, you can add targets which define the tools that your gateway will host. Gateway supports multiple target\ntypes including Lambda functions and API specifications (either OpenAPI schemas or Smithy models). Gateway allows you to attach multiple\ntargets to a Gateway and you can change the targets / tools attached to a gateway at any point. Each target can have its own\ncredential provider attached enabling you to securely access targets whether they need IAM, API Key, or OAuth credentials.\n\n### Gateway Target Properties\n\n| Name | Type | Required | Description |\n|------|------|----------|-------------|\n| `gatewayTargetName` | `string` | Yes | The name of the gateway target. Valid characters are a-z, A-Z, 0-9, _ (underscore) and - (hyphen) |\n| `description` | `string` | No | Optional description for the gateway target. Maximum 200 characters |\n| `gateway` | `IGateway` | Yes | The gateway this target belongs to |\n| `targetConfiguration` | `ITargetConfiguration` | Yes | The target configuration (Lambda, OpenAPI, or Smithy). **Note:** Users typically don't create this directly. When using convenience methods like `GatewayTarget.forLambda()`, `GatewayTarget.forOpenApi()`, `GatewayTarget.forSmithy()` or the gateway's `addLambdaTarget()`, `addOpenApiTarget()`, `addSmithyTarget()` methods, this configuration is created internally for you. Only needed when using the GatewayTarget constructor directly for [advanced scenarios](#advanced-usage-direct-configuration-for-gateway-target). |\n| `credentialProviderConfigurations` | `IGatewayCredentialProvider[]` | No | Credential providers for authentication. Defaults to `[GatewayCredentialProvider.fromIamRole()]`. Use `GatewayCredentialProvider.fromApiKeyIdentityArn()`, `GatewayCredentialProvider.fromOauthIdentityArn()`, or `GatewayCredentialProvider.fromIamRole()` |\n| `validateOpenApiSchema` | `boolean` | No | (OpenAPI targets only) Whether to validate the OpenAPI schema at synthesis time. Defaults to `true`. Only applies to inline and local asset schemas. For more information refer here <https://docs.aws.amazon.com/bedrock-agentcore/latest/devguide/gateway-schema-openapi.html> |\n\nThis approach gives you full control over the configuration but is typically not necessary for most use cases.\n\n### Targets types\n\nYou can create the following targets types:\n\n**Lambda Target**: Lambda targets allow you to connect your gateway to AWS Lambda functions that implement your tools. This is useful\nwhen you want to execute custom code in response to tool invocations.\n\n- Supports GATEWAY_IAM_ROLE credential provider only\n- Ideal for custom serverless function integration\n- Need tool schema (tool schema is a blueprint that describes the functions your Lambda provides to AI agents).\n  The construct provide [3 ways to upload a tool schema to Lambda target](#tools-schema-for-lambda-target)\n- When using the default IAM authentication (no `credentialProviderConfigurations` specified),\n  the construct automatocally grants the gateway role permission to invoke your Lambda function (`lambda:InvokeFunction`).\n\n**OpenAPI Schema Target** : OpenAPI widely used standard for describing RESTful APIs. Gateway supports OpenAPI 3.0\nspecifications for defining API targets. It connects to REST APIs using OpenAPI specifications\n\n- Supports OAUTH and API_KEY credential providers (Do not support IAM, you must provide `credentialProviderConfigurations`)\n- Ideal for integrating with external REST services\n- Need API schema. The construct provide [3 ways to upload a API schema to OpenAPI target](#api-schema-for-openapi-and-smithy-target)\n\n**Smithy Model Target** : Smithy is a language for defining services and software development kits (SDKs). Smithy models provide\na more structured approach to defining APIs compared to OpenAPI, and are particularly useful for connecting to AWS services.\nAgentCore Gateway supports built-in AWS service models only. It connects to services using Smithy model definitions\n\n- Supports OAUTH and API_KEY credential providers\n- Ideal for AWS service integrations\n- Need API schema. The construct provide 3 ways to upload a API schema to Smity target\n- When using the default IAM authentication (no `credentialProviderConfigurations` specified), The construct only\n  grants permission to read the Smithy schema file from S3. You MUST manually grant permissions for the gateway\n  role to invoke the actual Smithy API endpoints\n\n> Note: For Smithy model targets that access AWS services, your Gateway's execution role needs permissions to access those services.\nFor example, for a DynamoDB target, your execution role needs permissions to perform DynamoDB operations.\nThis is not managed by the construct due to the large number of options. Please refer to\n[Smithy Model Permission](https://docs.aws.amazon.com/bedrock-agentcore/latest/devguide/gateway-prerequisites-permissions.html) for example.\n\n**MCP Server Target**: Model Context Protocol (MCP) servers provide external tools, data access, and custom functions for AI agents.\nMCP servers enable agents to interact with external systems and services through a standardized protocol. Gateway automatically\ndiscovers and indexes available tools from MCP servers through synchronization.\n\n**Key Features:**\n\n- Requires explicit authentication configuration (OAuth2 recommended, empty array for NoAuth)\n- Ideal for connecting to external MCP-compliant servers\n- The endpoint must use HTTPS protocol\n- Supported MCP protocol versions: 2025-06-18, 2025-03-26\n- Automatic tool discovery through synchronization\n\n**Synchronization Behavior:**\n\nMCP Server targets require synchronization to discover and index available tools:\n\n- **Implicit Synchronization (Automatic)**: Tool discovery happens automatically during:\n  - Target creation (`CreateGatewayTarget`)\n  - Target updates (`UpdateGatewayTarget`)\n  - The Gateway calls the MCP server's `tools/list` endpoint and indexes tools without user intervention\n\n- **Explicit Synchronization (Manual)**: When the MCP server's tools change independently (new tools added, schemas modified, tools removed):\n  - The Gateway's tool catalog becomes stale\n  - Call the `SynchronizeGatewayTargets` API to refresh the catalog\n  - Use the `grantSync()` method to grant permissions to Lambda functions, CI/CD pipelines, or scheduled tasks that will trigger synchronization\n\n**Authentication & Permissions:**\n\nWhen using OAuth2, the Gateway service role automatically receives:\n\n- `bedrock-agentcore:GetWorkloadAccessToken`\n- `bedrock-agentcore:GetResourceOauth2Token`\n- `secretsmanager:GetSecretValue`\n- KMS decrypt (if secrets are encrypted)\n\nFor explicit synchronization, use `grantSync()` to grant `bedrock-agentcore:SynchronizeGatewayTargets` permission to your operator roles.\n\n> For more information, refer to the [MCP Server Target documentation](https://docs.aws.amazon.com/bedrock-agentcore/latest/devguide/gateway-target-MCPservers.html).\n\n### Understanding Tool Naming\n\nWhen tools are exposed through gateway targets, AgentCore Gateway prefixes each tool name with the target name to ensure uniqueness across multiple targets. This is important to understand when building your application logic.\n\n**Naming Pattern:**\n\n**Example:**\n\nIf your target is named `my-lambda-target` and provides a tool called `calculate_price`, agents will discover and invoke it as `my-lambda-target__calculate_price`.\n\n**Important Considerations:**\n\n- **For Lambda Targets**: Your Lambda handler must strip the target name prefix before processing the tool request. The full tool name (with prefix) is sent in the event.\n- **For MCP Server Targets**: The MCP server receives tool calls with the prefixed name from the gateway.\n- **For OpenAPI/Smithy Targets**: The gateway handles the prefix automatically when mapping to API operations based on the `operationId`.\n\nThis naming convention ensures that:\n\n- Tools from different targets don't collide even if they have the same name\n- Agents can access tools from multiple targets through a single gateway\n- Tool names remain unique in the unified tool catalog\n\nFor more details, see the [Gateway Tool Naming Documentation](https://docs.aws.amazon.com/bedrock-agentcore/latest/devguide/gateway-tool-naming.html).\n\n### Tools schema For Lambda target\n\nThe lambda target need tools schema to understand the fuunction lambda provides. You can upload the tool schema by following 3 ways:\n\n- From a local asset file\n\n```typescript\nconst toolSchema = agentcore.ToolSchema.fromLocalAsset(\n    path.join(__dirname, \"schemas\", \"my-tool-schema.json\")\n  );\n```\n\n- From an existing S3 file:\n\n```typescript\n\nconst toolSchema = agentcore.ToolSchema.fromS3File(\n    s3.Bucket.fromBucketName(this, \"SchemasBucket\", \"my-schemas-bucket\"),\n    \"tools/complex-tool-schema.json\",\n    \"123456789012\"\n  );\n```\n\n- From Inline:\n\n```typescript\nconst toolSchema = agentcore.ToolSchema.fromInline([{\n      name: \"hello_world\",\n      description: \"A simple hello world tool\",\n      inputSchema: {\n        type: agentcore.SchemaDefinitionType.OBJECT,\n        properties: {\n          name: {\n            type: agentcore.SchemaDefinitionType.STRING,\n            description: \"The name to greet\",\n          },\n        },\n        required: [\"name\"],\n      },\n    }]);\n\n```\n\n### Api schema For OpenAPI and Smithy target\n\nThe OpenAPI and Smithy target need API Schema. The Gateway construct provide three ways to upload API schema for your target:\n\n- From a local asset file (requires binding to scope):\n\n```typescript fixture=default\n// When using ApiSchema.fromLocalAsset, you must bind the schema to a scope\nconst schema = agentcore.ApiSchema.fromLocalAsset(path.join(__dirname, \"mySchema.yml\"));\n\nschema.bind(this);\n```\n\n- From an inline schema:\n\n```typescript fixture=default\nconst inlineSchema = agentcore.ApiSchema.fromInline(`\nopenapi: 3.0.3\ninfo:\n  title: Library API\n  version: 1.0.0\npaths:\n  /search:\n    get:\n      summary: Search for books\n      operationId: searchBooks\n      parameters:\n        - name: query\n          in: query\n          required: true\n          schema:\n            type: string\n`);\n```\n\n- From an existing S3 file:\n\n```typescript fixture=default\nconst bucket = s3.Bucket.fromBucketName(this, \"ExistingBucket\", \"my-schema-bucket\");\nconst s3Schema = agentcore.ApiSchema.fromS3File(bucket, \"schemas/action-group.yaml\");\n```\n\n### Outbound auth\n\nOutbound authorization lets Amazon Bedrock AgentCore gateways securely access gateway targets on behalf of users authenticated\nand authorized during Inbound Auth.\n\nAgentCore Gateway supports the following types of outbound authorization:\n\n**IAM-based outbound authorization** – The gateway uses its execution role to authenticate with AWS services. This is the default\n and most common approach for Lambda targets and AWS service integrations.\n\n**2-legged OAuth (OAuth 2LO)** – Use OAuth 2.0 two-legged flow (2LO) for targets that require OAuth authentication.\nThe gateway authenticates on its own behalf, not on behalf of a user.\n\n**API key** – Use the AgentCore service/AWS console to generate an API key to authenticate access to the gateway target.\n\n**Note > You need to set up the outbound identity before you can create a gateway target.\n\n### Basic Gateway Target Creation\n\nYou can create targets in two ways: using the static factory methods on `GatewayTarget` or using the convenient `addTarget` methods on the gateway instance.\n\n#### Using addTarget methods (Recommended)\n\nBelow are the examples on how you can create Lambda , Smity and OpenAPI target using `addTarget` method.\n\n```typescript fixture=default\n// Create a gateway first\nconst gateway = new agentcore.Gateway(this, \"MyGateway\", {\n  gatewayName: \"my-gateway\",\n});\n\nconst lambdaFunction = new lambda.Function(this, \"MyFunction\", {\n  runtime: lambda.Runtime.NODEJS_22_X,\n  handler: \"index.handler\",\n  code: lambda.Code.fromInline(`\n    exports.handler = async (event) => {\n      return {\n        statusCode: 200,\n        body: JSON.stringify({ message: 'Hello from Lambda!' })\n      };\n    };\n  `),\n});\n\nconst lambdaTarget = gateway.addLambdaTarget(\"MyLambdaTarget\", {\n  gatewayTargetName: \"my-lambda-target\",\n  description: \"Lambda function target\",\n  lambdaFunction: lambdaFunction,\n  toolSchema: agentcore.ToolSchema.fromInline([\n    {\n      name: \"hello_world\",\n      description: \"A simple hello world tool\",\n      inputSchema: {\n        type: agentcore.SchemaDefinitionType.OBJECT,\n        properties: {\n          name: {\n            type: agentcore.SchemaDefinitionType.STRING,\n            description: \"The name to greet\",\n          },\n        },\n        required: [\"name\"],\n      },\n    },\n  ]),\n});\n```\n\n- OpenAPI Target\n\n``` typescript\nconst gateway = new agentcore.Gateway(this, \"MyGateway\", {\n  gatewayName: \"my-gateway\",\n});\n\n// These ARNs are returned when creating the API key credential provider via Console or API\nconst apiKeyProviderArn = \"arn:aws:bedrock-agentcore:us-east-1:123456789012:token-vault/abc123/apikeycredentialprovider/my-apikey\"\nconst apiKeySecretArn = \"arn:aws:secretsmanager:us-east-1:123456789012:secret:my-apikey-secret-abc123\"\n\nconst bucket = s3.Bucket.fromBucketName(this, \"ExistingBucket\", \"my-schema-bucket\");\nconst s3mySchema = agentcore.ApiSchema.fromS3File(bucket, \"schemas/myschema.yaml\");\n\n// Add an OpenAPI target directly to the gateway\nconst target = gateway.addOpenApiTarget(\"MyTarget\", {\n  gatewayTargetName: \"my-api-target\",\n  description: \"Target for external API integration\",\n  apiSchema: s3mySchema,\n  credentialProviderConfigurations: [\n    agentcore.GatewayCredentialProvider.fromApiKeyIdentityArn({\n      providerArn: apiKeyProviderArn,\n      secretArn: apiKeySecretArn,\n      credentialLocation: agentcore.ApiKeyCredentialLocation.header({\n        credentialParameterName: \"X-API-Key\",\n      }),\n    }),\n  ],\n});\n\n// This make sure your s3 bucket is available before target\ntarget.node.addDependency(bucket);\n```\n\n- Smithy Target\n\n```typescript fixture=default\n\nconst gateway = new agentcore.Gateway(this, \"MyGateway\", {\n  gatewayName: \"my-gateway\",\n});\n\nconst smithySchema = agentcore.ApiSchema.fromLocalAsset(\n  path.join(__dirname, \"models\", \"smithy-model.json\")\n);\nsmithySchema.bind(this);\n\nconst smithyTarget = gateway.addSmithyTarget(\"MySmithyTarget\", {\n  gatewayTargetName: \"my-smithy-target\",\n  description: \"Smithy model target\",\n  smithyModel: smithySchema,\n\n});\n```\n\n- MCP Server Target\n\n```typescript fixture=default\nconst gateway = new agentcore.Gateway(this, \"MyGateway\", {\n  gatewayName: \"my-gateway\",\n});\n\n// OAuth2 authentication (recommended)\n// Note: Create the OAuth provider using AWS console or Identity L2 construct when available\nconst oauthProviderArn = \"arn:aws:bedrock-agentcore:us-east-1:123456789012:token-vault/abc123/oauth2credentialprovider/my-oauth\";\nconst oauthSecretArn = \"arn:aws:secretsmanager:us-east-1:123456789012:secret:my-oauth-secret-abc123\";\n\n// Add an MCP server target directly to the gateway\nconst mcpTarget = gateway.addMcpServerTarget(\"MyMcpServer\", {\n  gatewayTargetName: \"my-mcp-server\",\n  description: \"External MCP server integration\",\n  endpoint: \"https://my-mcp-server.example.com\",\n  credentialProviderConfigurations: [\n    agentcore.GatewayCredentialProvider.fromOauthIdentityArn({\n      providerArn: oauthProviderArn,\n      secretArn: oauthSecretArn,\n      scopes:['mcp-runtime-server/invoke']\n    }),\n  ],\n});\n\n// Grant sync permission to a Lambda function that will trigger synchronization\nconst syncFunction = new lambda.Function(this, \"SyncFunction\", {\n  runtime: lambda.Runtime.PYTHON_3_12,\n  handler: \"index.handler\",\n  code: lambda.Code.fromInline(`\nimport boto3\n\ndef handler(event, context):\n    client = boto3.client('bedrock-agentcore')\n    response = client.synchronize_gateway_targets(\n        gatewayIdentifier=event['gatewayId'],\n        targetIds=[event['targetId']]\n    )\n    return response\n  `),\n});\n\nmcpTarget.grantSync(syncFunction);\n```\n\n#### Using static factory methods\n\nCreate Gateway target using static convienence method.\n\n- Lambda Target\n\n```typescript fixture=default\nconst gateway = new agentcore.Gateway(this, \"MyGateway\", {\n  gatewayName: \"my-gateway\",\n});\n\nconst lambdaFunction = new lambda.Function(this, \"MyFunction\", {\n  runtime: lambda.Runtime.NODEJS_22_X,\n  handler: \"index.handler\",\n  code: lambda.Code.fromInline(`\n        exports.handler = async (event) => {\n            return {\n                statusCode: 200,\n                body: JSON.stringify({ message: 'Hello from Lambda!' })\n            };\n        };\n    `),\n});\n\n// Create a gateway target with Lambda and tool schema\nconst target = agentcore.GatewayTarget.forLambda(this, \"MyLambdaTarget\", {\n  gatewayTargetName: \"my-lambda-target\",\n  description: \"Target for Lambda function integration\",\n  gateway: gateway,\n  lambdaFunction: lambdaFunction,\n  toolSchema: agentcore.ToolSchema.fromLocalAsset(\n    path.join(__dirname, \"schemas\", \"my-tool-schema.json\")\n  ),\n});\n```\n\n- OpenAPI Target\n\n```typescript fixture=default\n\nconst gateway = new agentcore.Gateway(this, \"MyGateway\", {\n  gatewayName: \"my-gateway\",\n});\n\n// outbound auth (Use AWS console to create it, Once Identity L2 construct is available you can use it to create identity)\nconst apiKeyIdentityArn = \"arn:aws:bedrock-agentcore:us-east-1:123456789012:token-vault/abc123/apikeycredentialprovider/my-apikey\"\nconst apiKeySecretArn = \"arn:aws:secretsmanager:us-east-1:123456789012:secret:my-apikey-secret-abc123\"\n\nconst opneapiSchema = agentcore.ApiSchema.fromLocalAsset(path.join(__dirname, \"mySchema.yml\"));\nopneapiSchema.bind(this);\n\n// Create a gateway target with OpenAPI Schema\nconst target = agentcore.GatewayTarget.forOpenApi(this, \"MyTarget\", {\n  gatewayTargetName: \"my-api-target\",\n  description: \"Target for external API integration\",\n  gateway: gateway,  // Note: you need to pass the gateway reference\n  apiSchema: opneapiSchema,\n  credentialProviderConfigurations: [\n    agentcore.GatewayCredentialProvider.fromApiKeyIdentityArn({\n     providerArn: apiKeyIdentityArn,\n     secretArn: apiKeySecretArn\n    }),\n  ],\n});\n\n```\n\n- Smithy Target\n\n```typescript fixture=default\n\nconst gateway = new agentcore.Gateway(this, \"MyGateway\", {\n  gatewayName: \"my-gateway\",\n});\n\nconst smithySchema = agentcore.ApiSchema.fromLocalAsset(\n  path.join(__dirname, \"models\", \"smithy-model.json\")\n);\nsmithySchema.bind(this);\n\n// Create a gateway target with Smithy Model and OAuth\nconst target = agentcore.GatewayTarget.forSmithy(this, \"MySmithyTarget\", {\n  gatewayTargetName: \"my-smithy-target\",\n  description: \"Target for Smithy model integration\",\n  gateway: gateway,\n  smithyModel: smithySchema,\n});\n\n```\n\n- MCP Server Target\n\n```typescript fixture=default\nconst gateway = new agentcore.Gateway(this, \"MyGateway\", {\n  gatewayName: \"my-gateway\",\n});\n\n// OAuth2 authentication (recommended)\n// Note: Create the OAuth provider using AWS console or Identity L2 construct when available\nconst oauthProviderArn = \"arn:aws:bedrock-agentcore:us-east-1:123456789012:token-vault/abc123/oauth2credentialprovider/my-oauth\";\nconst oauthSecretArn = \"arn:aws:secretsmanager:us-east-1:123456789012:secret:my-oauth-secret-abc123\";\n\n// Create a gateway target with MCP Server\nconst mcpTarget = agentcore.GatewayTarget.forMcpServer(this, \"MyMcpServer\", {\n  gatewayTargetName: \"my-mcp-server\",\n  description: \"External MCP server integration\",\n  gateway: gateway,\n  endpoint: \"https://my-mcp-server.example.com\",\n  credentialProviderConfigurations: [\n    agentcore.GatewayCredentialProvider.fromOauthIdentityArn({\n      providerArn: oauthProviderArn,\n      secretArn: oauthSecretArn,\n      scopes:['mcp-runtime-server/invoke']\n    }),\n  ],\n});\n```\n\n### Advanced Usage: Direct Configuration for gateway target\n\nFor advanced use cases where you need full control over the target configuration, you can create configurations manually using the static factory methods and use the GatewayTarget constructor directly.\n\n#### Configuration Factory Methods\n\nEach target type has a corresponding configuration class with a static `create()` method:\n\n- **Lambda**: `LambdaTargetConfiguration.create(lambdaFunction, toolSchema)`\n- **OpenAPI**: `OpenApiTargetConfiguration.create(apiSchema, validateSchema?)`\n- **Smithy**: `SmithyTargetConfiguration.create(smithyModel)`\n\n#### Example: Lambda Target with Custom Configuration\n\n```typescript\nconst gateway = new agentcore.Gateway(this, \"MyGateway\", {\n  gatewayName: \"my-gateway\",\n});\n\nconst myLambdaFunction = new lambda.Function(this, \"MyFunction\", {\n  runtime: lambda.Runtime.NODEJS_22_X,\n  handler: \"index.handler\",\n  code: lambda.Code.fromInline(`\n    exports.handler = async (event) => ({ statusCode: 200 });\n  `),\n});\n\nconst myToolSchema = agentcore.ToolSchema.fromInline([{\n  name: \"my_tool\",\n  description: \"My custom tool\",\n  inputSchema: {\n    type: agentcore.SchemaDefinitionType.OBJECT,\n    properties: {},\n  },\n}]);\n\n// Create a custom Lambda configuration\nconst customConfig = agentcore.LambdaTargetConfiguration.create(\n  myLambdaFunction,\n  myToolSchema\n);\n\n// Use the GatewayTarget constructor directly\nconst target = new agentcore.GatewayTarget(this, \"AdvancedTarget\", {\n  gateway: gateway,\n  gatewayTargetName: \"advanced-target\",\n  targetConfiguration: customConfig,  // Manually created configuration\n  credentialProviderConfigurations: [\n    agentcore.GatewayCredentialProvider.fromIamRole()\n  ]\n});\n```\n\nThis approach gives you full control over the configuration but is typically not necessary for most use cases. The convenience methods (`GatewayTarget.forLambda()`, `GatewayTarget.forOpenApi()`, `GatewayTarget.forSmithy()`) handle all of this internally.\n\n### Gateway Target IAM Permissions\n\nThe Gateway Target construct provides convenient methods for granting IAM permissions:\n\n```typescript fixture=default\n// Create a gateway and target\nconst gateway = new agentcore.Gateway(this, \"MyGateway\", {\n  gatewayName: \"my-gateway\",\n});\n\nconst smithySchema = agentcore.ApiSchema.fromLocalAsset(\n  path.join(__dirname, \"models\", \"smithy-model.json\")\n);\nsmithySchema.bind(this);\n\n// Create a gateway target with Smithy Model and OAuth\nconst target = agentcore.GatewayTarget.forSmithy(this, \"MySmithyTarget\", {\n  gatewayTargetName: \"my-smithy-target\",\n  description: \"Target for Smithy model integration\",\n  gateway: gateway,\n  smithyModel: smithySchema,\n});\n\n// Create a role that needs access to the gateway target\nconst userRole = new iam.Role(this, \"UserRole\", {\n  assumedBy: new iam.ServicePrincipal(\"lambda.amazonaws.com\"),\n});\n\n// Grant read permissions (Get and List actions)\ntarget.grantRead(userRole);\n\n// Grant manage permissions (Create, Update, Delete actions)\ntarget.grantManage(userRole);\n\n// Grant specific custom permissions\ntarget.grant(userRole, \"bedrock-agentcore:GetGatewayTarget\");\n\n\n// Grants permission to invoke this Gateway\ngateway.grantInvoke(userRole);\n```\n\n## Memory\n\nMemory is a critical component of intelligence. While Large Language Models (LLMs) have impressive capabilities, they lack persistent memory across conversations. Amazon Bedrock AgentCore Memory addresses this limitation by providing a managed service that enables AI agents to maintain context over time, remember important facts, and deliver consistent, personalized experiences.\n\nAgentCore Memory operates on two levels:\n\n- **Short-Term Memory**: Immediate conversation context and session-based information that provides continuity within a single interaction or closely related sessions.\n- **Long-Term Memory**: Persistent information extracted and stored across multiple conversations, including facts, preferences, and summaries that enable personalized experiences over time.\n\nWhen you interact with the memory via the `CreateEvent` API, you store interactions in Short-Term Memory (STM) instantly. These interactions can include everything from user messages, assistant responses, to tool actions.\n\nTo write to long-term memory, you need to configure extraction strategies which define how and where to store information from conversations for future use. These strategies are asynchronously processed from raw events after every few turns based on the strategy that was selected. You can't create long term memory records directly, as they are extracted asynchronously by AgentCore Memory.\n\n### Memory Properties\n\n| Name | Type | Required | Description |\n|------|------|----------|-------------|\n| `memoryName` | `string` | Yes | The name of the memory |\n| `expirationDuration` | `Duration` | No | Short-term memory expiration in days (between 7 and 365). Default: 90 days |\n| `description` | `string` | No | Optional description for the memory. Default: no description. |\n| `kmsKey` | `IKey` | No | Custom KMS key to use for encryption. Default: Your data is encrypted with a key that AWS owns and manages for you |\n| `memoryStrategies` | `MemoryStrategyBase[]` | No | Built-in extraction strategies to use for this memory. Default: No extraction strategies (short term memory only) |\n| `executionRole` | `iam.IRole` | No | The IAM role that provides permissions for the memory to access AWS services. Default: A new role will be created. |\n| `tags` | `{ [key: string]: string }` | No | Tags for memory. Default: no tags. |\n\n### Basic Memory Creation\n\nBelow you can find how to configure a simple short-term memory (STM) with no long-term memory extraction strategies.\nNote how you set `expirationDuration`, which defines the time the events will be stored in the short-term memory before they expire.\n\n```typescript fixture=default\n\n// Create a basic memory with default settings, no LTM strategies\nconst memory = new agentcore.Memory(this, \"MyMemory\", {\n  memoryName: \"my_memory\",\n  description: \"A memory for storing user interactions for a period of 90 days\",\n  expirationDuration: cdk.Duration.days(90),\n});\n```\n\nBasic Memory with Custom KMS Encryption\n\n```typescript fixture=default\n// Create a custom KMS key for encryption\nconst encryptionKey = new kms.Key(this, \"MemoryEncryptionKey\", {\n  enableKeyRotation: true,\n  description: \"KMS key for memory encryption\",\n});\n\n// Create memory with custom encryption\nconst memory = new agentcore.Memory(this, \"MyMemory\", {\n  memoryName: \"my_encrypted_memory\",\n  description: \"Memory with custom KMS encryption\",\n  expirationDuration: cdk.Duration.days(90),\n  kmsKey: encryptionKey,\n});\n```\n\n### LTM Memory Extraction Stategies\n\nIf you need long-term memory for context recall across sessions, you can setup memory extraction strategies\nto extract the relevant memory from the raw events.\n\nAmazon Bedrock AgentCore Memory has different memory strategies for extracting and organizing information:\n\n- **Summarization**: to summarize interactions to preserve critical context and key insights.\n- **Semantic Memory**: to extract general factual knowledge, concepts and meanings from raw conversations using vector embeddings.\nThis enables similarity-based retrieval of relevant facts and context.\n- **User Preferences**: to extract user behavior patterns from raw conversations.\n\nYou can use built-in extraction strategies for quick setup, or create custom extraction strategies with specific models and prompt templates.\n\n### Memory with Built-in Strategies\n\nThe library provides three built-in LTM strategies. These are default strategies for organizing and extracting memory data,\neach optimized for specific use cases.\n\nFor example: An agent helps multiple users with cloud storage setup. From these conversations,\nsee how each strategy processes users expressing confusion about account connection:\n\n1. **Summarization Strategy** (`MemoryStrategy.usingBuiltInSummarization()`)\nThis strategy compresses conversations into concise overviews, preserving essential context and key insights for quick recall.\nExtracted memory example: Users confused by cloud setup during onboarding.\n\n   - Extracts concise summaries to preserve critical context and key insights\n   - Namespace: `/strategies/{memoryStrategyId}/actors/{actorId}/sessions/{sessionId}`\n\n2. **Semantic Memory Strategy** (`MemoryStrategy.usingBuiltInSemantic()`)\nDistills general facts, concepts, and underlying meanings from raw conversational data, presenting the information in a context-independent format.\nExtracted memory example: In-context learning = task-solving via examples, no training needed.\n\n   - Extracts general factual knowledge, concepts and meanings from raw conversations\n   - Namespace: `/strategies/{memoryStrategyId}/actors/{actorId}`\n\n3. **User Preference Strategy** (`MemoryStrategy.usingBuiltInUserPreference()`)\nCaptures individual preferences, interaction patterns, and personalized settings to enhance future experiences.\nExtracted memory example: User needs clear guidance on cloud storage account connection during onboarding.\n\n   - Extracts user behavior patterns from raw conversations\n   - Namespace: `/strategies/{memoryStrategyId}/actors/{actorId}`\n\n```typescript fixture=default\n// Create memory with built-in strategies\nconst memory = new agentcore.Memory(this, \"MyMemory\", {\n  memoryName: \"my_memory\",\n  description: \"Memory with built-in strategies\",\n  expirationDuration: cdk.Duration.days(90),\n  memoryStrategies: [\n    agentcore.MemoryStrategy.usingBuiltInSummarization(),\n    agentcore.MemoryStrategy.usingBuiltInSemantic(),\n    agentcore.MemoryStrategy.usingBuiltInUserPreference(),\n  ],\n});\n```\n\nThe name generated for each built in memory strategy is as follows:\n\n- For Summarization: `summary_builtin_cdk001`\n- For Semantic:`semantic_builtin_cdk001>`\n- For User Preferences: `preference_builtin_cdk001`\n\n### Memory with custom Strategies\n\nWith Long-Term Memory, organization is managed through Namespaces.\n\nAn `actor` refers to entity such as end users or agent/user combinations. For example, in a coding support chatbot,\nthe actor is usually the developer asking questions. Using the actor ID helps the system know which user the memory belongs to,\nkeeping each user's data separate and organized.\n\nA `session` is usually a single conversation or interaction period between the user and the AI agent.\nIt groups all related messages and events that happen during that conversation.\n\nA `namespace` is used to logically group and organize long-term memories. It ensures data stays neat, separate, and secure.\n\nWith AgentCore Memory, you need to add a namespace when you define a memory strategy. This namespace helps define where the long-term memory\nwill be logically grouped. Every time a new long-term memory is extracted using this memory strategy, it is saved under the namespace you set.\nThis means that all long-term memories are scoped to their specific namespace, keeping them organized and preventing any mix-ups with other\nusers or sessions. You should use a hierarchical format separated by forward slashes /. This helps keep memories organized clearly. As needed,\nyou can choose to use the below pre-defined variables within braces in the namespace based on your applications' organization needs:\n\n- `actorId` – Identifies who the long-term memory belongs to, such as a user\n- `memoryStrategyId` – Shows which memory strategy is being used. This strategy identifier is auto-generated when you create a memory using CreateMemory operation.\n- `sessionId` – Identifies which session or conversation the memory is from.\n\nFor example, if you define the following namespace as the input to your strategy in CreateMemory operation:\n\n```shell\n/strategy/{memoryStrategyId}/actor/{actorId}/session/{sessionId}\n```\n\nAfter memory creation, this namespace might look like:\n\n```shell\n/strategy/summarization-93483043//actor/actor-9830m2w3/session/session-9330sds8\n```\n\nYou can customise the namespace, i.e. where the memories are stored by using the following methods:\n\n1. **Summarization Strategy** (`MemoryStrategy.usingSummarization(props)`)\n1. **Semantic Memory Strategy** (`MemoryStrategy.usingSemantic(props)`)\n1. **User Preference Strategy** (`MemoryStrategy.usingUserPreference(props)`)\n\n```typescript fixture=default\n// Create memory with built-in strategies\nconst memory = new agentcore.Memory(this, \"MyMemory\", {\n  memoryName: \"my_memory\",\n  description: \"Memory with built-in strategies\",\n  expirationDuration: cdk.Duration.days(90),\n  memoryStrategies: [\n    agentcore.MemoryStrategy.usingUserPreference({\n        name: \"CustomerPreferences\",\n        namespaces: [\"support/customer/{actorId}/preferences\"]\n    }),\n    agentcore.MemoryStrategy.usingSemantic({\n        name: \"CustomerSupportSemantic\",\n        namespaces: [\"support/customer/{actorId}/semantic\"]\n    }),\n  ],\n});\n```\n\nCustom memory strategies let you tailor memory extraction and consolidation to your specific domain or use case.\nYou can override the prompts for extracting and consolidating semantic, summary, or user preferences.\nYou can also choose the model that you want to use for extraction and consolidation.\n\nThe custom prompts you create are appended to a non-editable system prompt.\n\nSince a custom strategy requires you to invoke certain FMs, you need a role with appropriate permissions. For that, you can:\n\n- Let the L2 construct create a minimum permission role for you when use L2 Bedrock Foundation Models.\n- Use a custom role with the overly permissive `AmazonBedrockAgentCoreMemoryBedrockModelInferenceExecutionRolePolicy` managed policy.\n- Use a custom role with your own custom policies.\n\n#### Memory with Custom Execution Role\n\nKeep in mind that memories that **do not** use custom strategies do not require a service role.\nSo even if you provide it, it will be ignored as it will never be used.\n\n```typescript fixture=default\n// Create a custom execution role\nconst executionRole = new iam.Role(this, \"MemoryExecutionRole\", {\n  assumedBy: new iam.ServicePrincipal(\"bedrock-agentcore.amazonaws.com\"),\n  managedPolicies: [\n    iam.ManagedPolicy.fromAwsManagedPolicyName(\n      \"AmazonBedrockAgentCoreMemoryBedrockModelInferenceExecutionRolePolicy\"\n    ),\n  ],\n});\n\n// Create memory with custom execution role\nconst memory = new agentcore.Memory(this, \"MyMemory\", {\n  memoryName: \"my_memory\",\n  description: \"Memory with custom execution role\",\n  expirationDuration: cdk.Duration.days(90),\n  executionRole: executionRole,\n});\n```\n\nIn customConsolidation and customExtraction, the model property uses the [@aws-cdk/aws-bedrock-alph](https://www.npmjs.com/package/@aws-cdk/aws-bedrock-alpha) library which must be installed separately.\n\n```typescript fixture=default\n// Create a custom semantic memory strategy\nconst customSemanticStrategy = agentcore.MemoryStrategy.usingSemantic({\n  name: \"customSemanticStrategy\",\n  description: \"Custom semantic memory strategy\",\n  namespaces: [\"/custom/strategies/{memoryStrategyId}/actors/{actorId}\"],\n  customConsolidation: {\n    model: bedrock.BedrockFoundationModel.ANTHROPIC_CLAUDE_3_5_SONNET_V1_0,\n    appendToPrompt: \"Custom consolidation prompt for semantic memory\",\n  },\n  customExtraction: {\n    model: bedrock.BedrockFoundationModel.ANTHROPIC_CLAUDE_3_5_SONNET_V1_0,\n    appendToPrompt: \"Custom extraction prompt for semantic memory\",\n  },\n});\n\n// Create memory with custom strategy\nconst memory = new agentcore.Memory(this, \"MyMemory\", {\n  memoryName: \"my-custom-memory\",\n  description: \"Memory with custom strategy\",\n  expirationDuration: cdk.Duration.days(90),\n  memoryStrategies: [customSemanticStrategy],\n});\n```\n\n### Memory with self-managed Strategies\n\nA self-managed strategy in Amazon Bedrock AgentCore Memory gives you complete control over your memory extraction and consolidation pipelines.\nWith a self-managed strategy, you can build custom memory processing workflows while leveraging Amazon Bedrock AgentCore for storage and retrieval.\n\nFor additional information, you can refer to the [developer guide for self managed strategies](https://docs.aws.amazon.com/bedrock-agentcore/latest/devguide/memory-self-managed-strategies.html).\n\nCreate the required AWS resources including:\n\n- an S3 bucket in your account where Amazon Bedrock AgentCore will deliver batched event payloads.\n- an SNS topic for job notifications. Use FIFO topics if processing order within sessions is important for your use case.\n\nThe construct will apply the correct permissions to the memory execution role to access these resources.\n\n```typescript fixture=default\n\nconst bucket = new s3.Bucket(this, 'memoryBucket', {\n  bucketName: 'test-memory',\n  removalPolicy: cdk.RemovalPolicy.DESTROY,\n  autoDeleteObjects: true,\n});\n\nconst topic = new sns.Topic(this, 'topic');\n\n// Create a custom semantic memory strategy\nconst selfManagedStrategy = agentcore.MemoryStrategy.usingSelfManaged({\n  name: \"selfManagedStrategy\",\n  description: \"self managed memory strategy\",\n  historicalContextWindowSize: 5,\n  invocationConfiguration: {\n    topic: topic,\n    s3Location: {\n      bucketName: bucket.bucketName,\n      objectKey: 'memory/',\n    }\n  },\n  triggerConditions: {\n    messageBasedTrigger: 1,\n    timeBasedTrigger: cdk.Duration.seconds(10),\n    tokenBasedTrigger: 100\n  }\n});\n\n// Create memory with custom strategy\nconst memory = new agentcore.Memory(this, \"MyMemory\", {\n  memoryName: \"my-custom-memory\",\n  description: \"Memory with custom strategy\",\n  expirationDuration: cdk.Duration.days(90),\n  memoryStrategies: [selfManagedStrategy],\n});\n```\n\n### Memory Strategy Methods\n\nYou can add new memory strategies to the memory construct using the `addMemoryStrategy()` method, for instance:\n\n```typescript fixture=default\n// Create memory without initial strategies\nconst memory = new agentcore.Memory(this, \"test-memory\", {\n  memoryName: \"test_memory_add_strategy\",\n  description: \"A test memory for testing addMemoryStrategy method\",\n  expirationDuration: cdk.Duration.days(90),\n});\n\n// Add strategies after instantiation\nmemory.addMemoryStrategy(agentcore.MemoryStrategy.usingBuiltInSummarization());\nmemory.addMemoryStrategy(agentcore.MemoryStrategy.usingBuiltInSemantic());\n```\n"
   },
   "repository": {
     "directory": "packages/@aws-cdk/aws-bedrock-agentcore-alpha",
@@ -9304,7 +9304,7 @@
       "kind": "interface",
       "locationInModule": {
         "filename": "lib/runtime/runtime-base.ts",
-        "line": 392
+        "line": 403
       },
       "name": "AgentRuntimeAttributes",
       "properties": [
@@ -9317,7 +9317,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "lib/runtime/runtime-base.ts",
-            "line": 396
+            "line": 407
           },
           "name": "agentRuntimeArn",
           "type": {
@@ -9333,7 +9333,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "lib/runtime/runtime-base.ts",
-            "line": 401
+            "line": 412
           },
           "name": "agentRuntimeId",
           "type": {
@@ -9349,7 +9349,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "lib/runtime/runtime-base.ts",
-            "line": 406
+            "line": 417
           },
           "name": "agentRuntimeName",
           "type": {
@@ -9365,7 +9365,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "lib/runtime/runtime-base.ts",
-            "line": 411
+            "line": 422
           },
           "name": "roleArn",
           "type": {
@@ -9382,7 +9382,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "lib/runtime/runtime-base.ts",
-            "line": 419
+            "line": 430
           },
           "name": "agentRuntimeVersion",
           "optional": true,
@@ -9400,7 +9400,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "lib/runtime/runtime-base.ts",
-            "line": 437
+            "line": 448
           },
           "name": "agentStatus",
           "optional": true,
@@ -9418,7 +9418,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "lib/runtime/runtime-base.ts",
-            "line": 443
+            "line": 454
           },
           "name": "createdAt",
           "optional": true,
@@ -9436,7 +9436,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "lib/runtime/runtime-base.ts",
-            "line": 425
+            "line": 436
           },
           "name": "description",
           "optional": true,
@@ -9454,7 +9454,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "lib/runtime/runtime-base.ts",
-            "line": 449
+            "line": 460
           },
           "name": "lastUpdatedAt",
           "optional": true,
@@ -9472,7 +9472,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "lib/runtime/runtime-base.ts",
-            "line": 431
+            "line": 442
           },
           "name": "securityGroups",
           "optional": true,
@@ -10206,7 +10206,7 @@
         },
         "locationInModule": {
           "filename": "lib/tools/browser.ts",
-          "line": 727
+          "line": 734
         },
         "parameters": [
           {
@@ -10232,7 +10232,7 @@
       "kind": "class",
       "locationInModule": {
         "filename": "lib/tools/browser.ts",
-        "line": 611
+        "line": 618
       },
       "methods": [
         {
@@ -10243,7 +10243,7 @@
           },
           "locationInModule": {
             "filename": "lib/tools/browser.ts",
-            "line": 627
+            "line": 634
           },
           "name": "fromBrowserCustomAttributes",
           "parameters": [
@@ -10294,7 +10294,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "lib/tools/browser.ts",
-            "line": 614
+            "line": 621
           },
           "name": "PROPERTY_INJECTION_ID",
           "static": true,
@@ -10313,7 +10313,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "lib/tools/browser.ts",
-            "line": 659
+            "line": 666
           },
           "name": "browserArn",
           "overrides": "@aws-cdk/aws-bedrock-agentcore-alpha.BrowserCustomBase",
@@ -10332,7 +10332,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "lib/tools/browser.ts",
-            "line": 664
+            "line": 671
           },
           "name": "browserId",
           "overrides": "@aws-cdk/aws-bedrock-agentcore-alpha.BrowserCustomBase",
@@ -10348,7 +10348,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "lib/tools/browser.ts",
-            "line": 696
+            "line": 703
           },
           "name": "executionRole",
           "overrides": "@aws-cdk/aws-bedrock-agentcore-alpha.BrowserCustomBase",
@@ -10364,7 +10364,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "lib/tools/browser.ts",
-            "line": 706
+            "line": 713
           },
           "name": "grantPrincipal",
           "overrides": "@aws-cdk/aws-bedrock-agentcore-alpha.BrowserCustomBase",
@@ -10380,7 +10380,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "lib/tools/browser.ts",
-            "line": 668
+            "line": 675
           },
           "name": "name",
           "type": {
@@ -10395,7 +10395,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "lib/tools/browser.ts",
-            "line": 710
+            "line": 717
           },
           "name": "networkConfiguration",
           "type": {
@@ -10410,7 +10410,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "lib/tools/browser.ts",
-            "line": 718
+            "line": 725
           },
           "name": "browserSigning",
           "optional": true,
@@ -10429,7 +10429,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "lib/tools/browser.ts",
-            "line": 687
+            "line": 694
           },
           "name": "createdAt",
           "optional": true,
@@ -10446,7 +10446,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "lib/tools/browser.ts",
-            "line": 672
+            "line": 679
           },
           "name": "description",
           "optional": true,
@@ -10465,7 +10465,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "lib/tools/browser.ts",
-            "line": 692
+            "line": 699
           },
           "name": "failureReason",
           "optional": true,
@@ -10484,7 +10484,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "lib/tools/browser.ts",
-            "line": 677
+            "line": 684
           },
           "name": "lastUpdatedAt",
           "optional": true,
@@ -10501,7 +10501,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "lib/tools/browser.ts",
-            "line": 714
+            "line": 721
           },
           "name": "recordingConfig",
           "optional": true,
@@ -10520,7 +10520,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "lib/tools/browser.ts",
-            "line": 682
+            "line": 689
           },
           "name": "status",
           "optional": true,
@@ -10538,7 +10538,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "lib/tools/browser.ts",
-            "line": 702
+            "line": 709
           },
           "name": "tags",
           "optional": true,
@@ -10569,7 +10569,7 @@
       "kind": "interface",
       "locationInModule": {
         "filename": "lib/tools/browser.ts",
-        "line": 567
+        "line": 574
       },
       "name": "BrowserCustomAttributes",
       "properties": [
@@ -10585,7 +10585,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "lib/tools/browser.ts",
-            "line": 572
+            "line": 579
           },
           "name": "browserArn",
           "type": {
@@ -10604,7 +10604,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "lib/tools/browser.ts",
-            "line": 577
+            "line": 584
           },
           "name": "roleArn",
           "type": {
@@ -10621,7 +10621,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "lib/tools/browser.ts",
-            "line": 592
+            "line": 599
           },
           "name": "createdAt",
           "optional": true,
@@ -10639,7 +10639,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "lib/tools/browser.ts",
-            "line": 582
+            "line": 589
           },
           "name": "lastUpdatedAt",
           "optional": true,
@@ -10657,7 +10657,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "lib/tools/browser.ts",
-            "line": 598
+            "line": 605
           },
           "name": "securityGroups",
           "optional": true,
@@ -10680,7 +10680,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "lib/tools/browser.ts",
-            "line": 587
+            "line": 594
           },
           "name": "status",
           "optional": true,
@@ -10735,13 +10735,14 @@
       "methods": [
         {
           "docs": {
+            "remarks": "[disable-awslint:no-grants]",
             "returns": "An IAM Grant object representing the granted permissions",
             "stability": "experimental",
             "summary": "Grants IAM actions to the IAM Principal."
           },
           "locationInModule": {
             "filename": "lib/tools/browser.ts",
-            "line": 242
+            "line": 245
           },
           "name": "grant",
           "overrides": "@aws-cdk/aws-bedrock-agentcore-alpha.IBrowserCustom",
@@ -10776,14 +10777,14 @@
         {
           "docs": {
             "default": "- Default grant configuration:\n- actions: ['bedrock-agentcore:GetBrowser', 'bedrock-agentcore:GetBrowserSession'] on this.browserArn\n- actions: ['bedrock-agentcore:ListBrowsers', 'bedrock-agentcore:ListBrowserSessions'] on all resources (*)",
-            "remarks": "This includes both read permissions on the specific browser and list permissions on all browsers.",
+            "remarks": "This includes both read permissions on the specific browser and list permissions on all browsers.\n\n[disable-awslint:no-grants]",
             "returns": "An IAM Grant object representing the granted permissions",
             "stability": "experimental",
             "summary": "Grant read permissions on this browser to an IAM principal."
           },
           "locationInModule": {
             "filename": "lib/tools/browser.ts",
-            "line": 260
+            "line": 265
           },
           "name": "grantRead",
           "overrides": "@aws-cdk/aws-bedrock-agentcore-alpha.IBrowserCustom",
@@ -10807,13 +10808,14 @@
         {
           "docs": {
             "default": "- Default grant configuration:\n- actions: ['bedrock-agentcore:StartBrowserSession', 'bedrock-agentcore:UpdateBrowserStream', 'bedrock-agentcore:StopBrowserSession']\n- resourceArns: [this.browserArn]",
+            "remarks": "[disable-awslint:no-grants]",
             "returns": "An IAM Grant object representing the granted permissions",
             "stability": "experimental",
             "summary": "Grant invoke permissions on this browser to an IAM principal."
           },
           "locationInModule": {
             "filename": "lib/tools/browser.ts",
-            "line": 284
+            "line": 291
           },
           "name": "grantUse",
           "overrides": "@aws-cdk/aws-bedrock-agentcore-alpha.IBrowserCustom",
@@ -10842,7 +10844,7 @@
           },
           "locationInModule": {
             "filename": "lib/tools/browser.ts",
-            "line": 300
+            "line": 307
           },
           "name": "metric",
           "overrides": "@aws-cdk/aws-bedrock-agentcore-alpha.IBrowserCustom",
@@ -10887,7 +10889,7 @@
           },
           "locationInModule": {
             "filename": "lib/tools/browser.ts",
-            "line": 366
+            "line": 373
           },
           "name": "metricErrorsForApiOperation",
           "overrides": "@aws-cdk/aws-bedrock-agentcore-alpha.IBrowserCustom",
@@ -10924,7 +10926,7 @@
           },
           "locationInModule": {
             "filename": "lib/tools/browser.ts",
-            "line": 320
+            "line": 327
           },
           "name": "metricForApiOperation",
           "overrides": "@aws-cdk/aws-bedrock-agentcore-alpha.IBrowserCustom",
@@ -10967,7 +10969,7 @@
           },
           "locationInModule": {
             "filename": "lib/tools/browser.ts",
-            "line": 350
+            "line": 357
           },
           "name": "metricInvocationsForApiOperation",
           "overrides": "@aws-cdk/aws-bedrock-agentcore-alpha.IBrowserCustom",
@@ -11004,7 +11006,7 @@
           },
           "locationInModule": {
             "filename": "lib/tools/browser.ts",
-            "line": 337
+            "line": 344
           },
           "name": "metricLatencyForApiOperation",
           "overrides": "@aws-cdk/aws-bedrock-agentcore-alpha.IBrowserCustom",
@@ -11041,7 +11043,7 @@
           },
           "locationInModule": {
             "filename": "lib/tools/browser.ts",
-            "line": 418
+            "line": 425
           },
           "name": "metricSessionDuration",
           "overrides": "@aws-cdk/aws-bedrock-agentcore-alpha.IBrowserCustom",
@@ -11072,7 +11074,7 @@
           },
           "locationInModule": {
             "filename": "lib/tools/browser.ts",
-            "line": 392
+            "line": 399
           },
           "name": "metricSystemErrorsForApiOperation",
           "overrides": "@aws-cdk/aws-bedrock-agentcore-alpha.IBrowserCustom",
@@ -11109,7 +11111,7 @@
           },
           "locationInModule": {
             "filename": "lib/tools/browser.ts",
-            "line": 431
+            "line": 438
           },
           "name": "metricTakeOverCount",
           "overrides": "@aws-cdk/aws-bedrock-agentcore-alpha.IBrowserCustom",
@@ -11140,7 +11142,7 @@
           },
           "locationInModule": {
             "filename": "lib/tools/browser.ts",
-            "line": 457
+            "line": 464
           },
           "name": "metricTakeOverDuration",
           "overrides": "@aws-cdk/aws-bedrock-agentcore-alpha.IBrowserCustom",
@@ -11171,7 +11173,7 @@
           },
           "locationInModule": {
             "filename": "lib/tools/browser.ts",
-            "line": 444
+            "line": 451
           },
           "name": "metricTakeOverReleaseCount",
           "overrides": "@aws-cdk/aws-bedrock-agentcore-alpha.IBrowserCustom",
@@ -11202,7 +11204,7 @@
           },
           "locationInModule": {
             "filename": "lib/tools/browser.ts",
-            "line": 379
+            "line": 386
           },
           "name": "metricThrottlesForApiOperation",
           "overrides": "@aws-cdk/aws-bedrock-agentcore-alpha.IBrowserCustom",
@@ -11239,7 +11241,7 @@
           },
           "locationInModule": {
             "filename": "lib/tools/browser.ts",
-            "line": 405
+            "line": 412
           },
           "name": "metricUserErrorsForApiOperation",
           "overrides": "@aws-cdk/aws-bedrock-agentcore-alpha.IBrowserCustom",
@@ -11426,7 +11428,7 @@
       "kind": "interface",
       "locationInModule": {
         "filename": "lib/tools/browser.ts",
-        "line": 503
+        "line": 510
       },
       "name": "BrowserCustomProps",
       "properties": [
@@ -11442,7 +11444,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "lib/tools/browser.ts",
-            "line": 511
+            "line": 518
           },
           "name": "browserCustomName",
           "type": {
@@ -11460,7 +11462,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "lib/tools/browser.ts",
-            "line": 558
+            "line": 565
           },
           "name": "browserSigning",
           "optional": true,
@@ -11481,7 +11483,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "lib/tools/browser.ts",
-            "line": 520
+            "line": 527
           },
           "name": "description",
           "optional": true,
@@ -11502,7 +11504,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "lib/tools/browser.ts",
-            "line": 541
+            "line": 548
           },
           "name": "executionRole",
           "optional": true,
@@ -11523,7 +11525,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "lib/tools/browser.ts",
-            "line": 527
+            "line": 534
           },
           "name": "networkConfiguration",
           "optional": true,
@@ -11544,7 +11546,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "lib/tools/browser.ts",
-            "line": 534
+            "line": 541
           },
           "name": "recordingConfig",
           "optional": true,
@@ -11565,7 +11567,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "lib/tools/browser.ts",
-            "line": 550
+            "line": 557
           },
           "name": "tags",
           "optional": true,
@@ -11750,7 +11752,7 @@
         },
         "locationInModule": {
           "filename": "lib/tools/code-interpreter.ts",
-          "line": 621
+          "line": 630
         },
         "parameters": [
           {
@@ -11776,7 +11778,7 @@
       "kind": "class",
       "locationInModule": {
         "filename": "lib/tools/code-interpreter.ts",
-        "line": 515
+        "line": 524
       },
       "methods": [
         {
@@ -11787,7 +11789,7 @@
           },
           "locationInModule": {
             "filename": "lib/tools/code-interpreter.ts",
-            "line": 531
+            "line": 540
           },
           "name": "fromCodeInterpreterCustomAttributes",
           "parameters": [
@@ -11838,7 +11840,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "lib/tools/code-interpreter.ts",
-            "line": 518
+            "line": 527
           },
           "name": "PROPERTY_INJECTION_ID",
           "static": true,
@@ -11857,7 +11859,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "lib/tools/code-interpreter.ts",
-            "line": 563
+            "line": 572
           },
           "name": "codeInterpreterArn",
           "overrides": "@aws-cdk/aws-bedrock-agentcore-alpha.CodeInterpreterCustomBase",
@@ -11876,7 +11878,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "lib/tools/code-interpreter.ts",
-            "line": 568
+            "line": 577
           },
           "name": "codeInterpreterId",
           "overrides": "@aws-cdk/aws-bedrock-agentcore-alpha.CodeInterpreterCustomBase",
@@ -11892,7 +11894,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "lib/tools/code-interpreter.ts",
-            "line": 604
+            "line": 613
           },
           "name": "executionRole",
           "overrides": "@aws-cdk/aws-bedrock-agentcore-alpha.CodeInterpreterCustomBase",
@@ -11908,7 +11910,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "lib/tools/code-interpreter.ts",
-            "line": 608
+            "line": 617
           },
           "name": "grantPrincipal",
           "overrides": "@aws-cdk/aws-bedrock-agentcore-alpha.CodeInterpreterCustomBase",
@@ -11924,7 +11926,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "lib/tools/code-interpreter.ts",
-            "line": 572
+            "line": 581
           },
           "name": "name",
           "type": {
@@ -11939,7 +11941,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "lib/tools/code-interpreter.ts",
-            "line": 580
+            "line": 589
           },
           "name": "networkConfiguration",
           "type": {
@@ -11957,7 +11959,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "lib/tools/code-interpreter.ts",
-            "line": 590
+            "line": 599
           },
           "name": "createdAt",
           "optional": true,
@@ -11974,7 +11976,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "lib/tools/code-interpreter.ts",
-            "line": 576
+            "line": 585
           },
           "name": "description",
           "optional": true,
@@ -11993,7 +11995,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "lib/tools/code-interpreter.ts",
-            "line": 600
+            "line": 609
           },
           "name": "failureReason",
           "optional": true,
@@ -12012,7 +12014,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "lib/tools/code-interpreter.ts",
-            "line": 595
+            "line": 604
           },
           "name": "lastUpdatedAt",
           "optional": true,
@@ -12032,7 +12034,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "lib/tools/code-interpreter.ts",
-            "line": 585
+            "line": 594
           },
           "name": "status",
           "optional": true,
@@ -12050,7 +12052,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "lib/tools/code-interpreter.ts",
-            "line": 614
+            "line": 623
           },
           "name": "tags",
           "optional": true,
@@ -12081,7 +12083,7 @@
       "kind": "interface",
       "locationInModule": {
         "filename": "lib/tools/code-interpreter.ts",
-        "line": 471
+        "line": 480
       },
       "name": "CodeInterpreterCustomAttributes",
       "properties": [
@@ -12097,7 +12099,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "lib/tools/code-interpreter.ts",
-            "line": 476
+            "line": 485
           },
           "name": "codeInterpreterArn",
           "type": {
@@ -12116,7 +12118,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "lib/tools/code-interpreter.ts",
-            "line": 481
+            "line": 490
           },
           "name": "roleArn",
           "type": {
@@ -12133,7 +12135,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "lib/tools/code-interpreter.ts",
-            "line": 496
+            "line": 505
           },
           "name": "createdAt",
           "optional": true,
@@ -12151,7 +12153,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "lib/tools/code-interpreter.ts",
-            "line": 486
+            "line": 495
           },
           "name": "lastUpdatedAt",
           "optional": true,
@@ -12169,7 +12171,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "lib/tools/code-interpreter.ts",
-            "line": 502
+            "line": 511
           },
           "name": "securityGroups",
           "optional": true,
@@ -12192,7 +12194,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "lib/tools/code-interpreter.ts",
-            "line": 491
+            "line": 500
           },
           "name": "status",
           "optional": true,
@@ -12247,13 +12249,14 @@
       "methods": [
         {
           "docs": {
+            "remarks": "[disable-awslint:no-grants]",
             "returns": "An IAM Grant object representing the granted permissions",
             "stability": "experimental",
             "summary": "Grants IAM actions to the IAM Principal."
           },
           "locationInModule": {
             "filename": "lib/tools/code-interpreter.ts",
-            "line": 207
+            "line": 210
           },
           "name": "grant",
           "overrides": "@aws-cdk/aws-bedrock-agentcore-alpha.ICodeInterpreterCustom",
@@ -12288,13 +12291,14 @@
         {
           "docs": {
             "default": "- Default grant configuration:\n- actions: ['bedrock-agentcore:InvokeCodeInterpreter']\n- resourceArns: [this.codeInterpreterArn]",
+            "remarks": "[disable-awslint:no-grants]",
             "returns": "An IAM Grant object representing the granted permissions",
             "stability": "experimental",
             "summary": "Grant invoke permissions on this code interpreter to an IAM principal."
           },
           "locationInModule": {
             "filename": "lib/tools/code-interpreter.ts",
-            "line": 265
+            "line": 274
           },
           "name": "grantInvoke",
           "parameters": [
@@ -12317,14 +12321,14 @@
         {
           "docs": {
             "default": "- Default grant configuration:\n- actions: ['bedrock-agentcore:GetCodeInterpreter', 'bedrock-agentcore:GetCodeInterpreterSession'] on this.codeInterpreterArn\n- actions: ['bedrock-agentcore:ListCodeInterpreters', 'bedrock-agentcore:ListCodeInterpreterSessions'] on all resources (*)",
-            "remarks": "This includes both read permissions on the specific code interpreter and list permissions on all code interpreters.",
+            "remarks": "This includes both read permissions on the specific code interpreter and list permissions on all code interpreters.\n\n[disable-awslint:no-grants]",
             "returns": "An IAM Grant object representing the granted permissions",
             "stability": "experimental",
             "summary": "Grant read permissions on this code interpreter to an IAM principal."
           },
           "locationInModule": {
             "filename": "lib/tools/code-interpreter.ts",
-            "line": 225
+            "line": 230
           },
           "name": "grantRead",
           "overrides": "@aws-cdk/aws-bedrock-agentcore-alpha.ICodeInterpreterCustom",
@@ -12348,13 +12352,14 @@
         {
           "docs": {
             "default": "- Default grant configuration:\n- actions: ['bedrock-agentcore:StartCodeInterpreterSession', 'bedrock-agentcore:InvokeCodeInterpreter', 'bedrock-agentcore:StopCodeInterpreterSession']\n- resourceArns: [this.codeInterpreterArn]",
+            "remarks": "[disable-awslint:no-grants]",
             "returns": "An IAM Grant object representing the granted permissions",
             "stability": "experimental",
             "summary": "Grant invoke permissions on this code interpreter to an IAM principal."
           },
           "locationInModule": {
             "filename": "lib/tools/code-interpreter.ts",
-            "line": 249
+            "line": 256
           },
           "name": "grantUse",
           "overrides": "@aws-cdk/aws-bedrock-agentcore-alpha.ICodeInterpreterCustom",
@@ -12383,7 +12388,7 @@
           },
           "locationInModule": {
             "filename": "lib/tools/code-interpreter.ts",
-            "line": 278
+            "line": 287
           },
           "name": "metric",
           "overrides": "@aws-cdk/aws-bedrock-agentcore-alpha.ICodeInterpreterCustom",
@@ -12428,7 +12433,7 @@
           },
           "locationInModule": {
             "filename": "lib/tools/code-interpreter.ts",
-            "line": 344
+            "line": 353
           },
           "name": "metricErrorsForApiOperation",
           "overrides": "@aws-cdk/aws-bedrock-agentcore-alpha.ICodeInterpreterCustom",
@@ -12465,7 +12470,7 @@
           },
           "locationInModule": {
             "filename": "lib/tools/code-interpreter.ts",
-            "line": 298
+            "line": 307
           },
           "name": "metricForApiOperation",
           "overrides": "@aws-cdk/aws-bedrock-agentcore-alpha.ICodeInterpreterCustom",
@@ -12508,7 +12513,7 @@
           },
           "locationInModule": {
             "filename": "lib/tools/code-interpreter.ts",
-            "line": 328
+            "line": 337
           },
           "name": "metricInvocationsForApiOperation",
           "overrides": "@aws-cdk/aws-bedrock-agentcore-alpha.ICodeInterpreterCustom",
@@ -12545,7 +12550,7 @@
           },
           "locationInModule": {
             "filename": "lib/tools/code-interpreter.ts",
-            "line": 315
+            "line": 324
           },
           "name": "metricLatencyForApiOperation",
           "overrides": "@aws-cdk/aws-bedrock-agentcore-alpha.ICodeInterpreterCustom",
@@ -12582,7 +12587,7 @@
           },
           "locationInModule": {
             "filename": "lib/tools/code-interpreter.ts",
-            "line": 396
+            "line": 405
           },
           "name": "metricSessionDuration",
           "overrides": "@aws-cdk/aws-bedrock-agentcore-alpha.ICodeInterpreterCustom",
@@ -12613,7 +12618,7 @@
           },
           "locationInModule": {
             "filename": "lib/tools/code-interpreter.ts",
-            "line": 370
+            "line": 379
           },
           "name": "metricSystemErrorsForApiOperation",
           "overrides": "@aws-cdk/aws-bedrock-agentcore-alpha.ICodeInterpreterCustom",
@@ -12650,7 +12655,7 @@
           },
           "locationInModule": {
             "filename": "lib/tools/code-interpreter.ts",
-            "line": 357
+            "line": 366
           },
           "name": "metricThrottlesForApiOperation",
           "overrides": "@aws-cdk/aws-bedrock-agentcore-alpha.ICodeInterpreterCustom",
@@ -12687,7 +12692,7 @@
           },
           "locationInModule": {
             "filename": "lib/tools/code-interpreter.ts",
-            "line": 383
+            "line": 392
           },
           "name": "metricUserErrorsForApiOperation",
           "overrides": "@aws-cdk/aws-bedrock-agentcore-alpha.ICodeInterpreterCustom",
@@ -12874,7 +12879,7 @@
       "kind": "interface",
       "locationInModule": {
         "filename": "lib/tools/code-interpreter.ts",
-        "line": 421
+        "line": 430
       },
       "name": "CodeInterpreterCustomProps",
       "properties": [
@@ -12890,7 +12895,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "lib/tools/code-interpreter.ts",
-            "line": 429
+            "line": 438
           },
           "name": "codeInterpreterCustomName",
           "type": {
@@ -12910,7 +12915,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "lib/tools/code-interpreter.ts",
-            "line": 438
+            "line": 447
           },
           "name": "description",
           "optional": true,
@@ -12931,7 +12936,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "lib/tools/code-interpreter.ts",
-            "line": 446
+            "line": 455
           },
           "name": "executionRole",
           "optional": true,
@@ -12952,7 +12957,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "lib/tools/code-interpreter.ts",
-            "line": 453
+            "line": 462
           },
           "name": "networkConfiguration",
           "optional": true,
@@ -12973,7 +12978,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "lib/tools/code-interpreter.ts",
-            "line": 462
+            "line": 471
           },
           "name": "tags",
           "optional": true,
@@ -13403,7 +13408,7 @@
         },
         "locationInModule": {
           "filename": "lib/gateway/gateway.ts",
-          "line": 392
+          "line": 414
         },
         "parameters": [
           {
@@ -13487,7 +13492,7 @@
           },
           "locationInModule": {
             "filename": "lib/gateway/gateway.ts",
-            "line": 454
+            "line": 483
           },
           "name": "addLambdaTarget",
           "parameters": [
@@ -13525,7 +13530,7 @@
           },
           "locationInModule": {
             "filename": "lib/gateway/gateway.ts",
-            "line": 540
+            "line": 569
           },
           "name": "addMcpServerTarget",
           "parameters": [
@@ -13562,7 +13567,7 @@
           },
           "locationInModule": {
             "filename": "lib/gateway/gateway.ts",
-            "line": 485
+            "line": 514
           },
           "name": "addOpenApiTarget",
           "parameters": [
@@ -13599,7 +13604,7 @@
           },
           "locationInModule": {
             "filename": "lib/gateway/gateway.ts",
-            "line": 510
+            "line": 539
           },
           "name": "addSmithyTarget",
           "parameters": [
@@ -13841,6 +13846,28 @@
             "fqn": "aws-cdk-lib.aws_kms.IKey"
           }
         },
+        {
+          "docs": {
+            "remarks": "Only available when using the default Cognito authorizer.",
+            "stability": "experimental",
+            "summary": "The OAuth2 scope strings for client credentials flow."
+          },
+          "immutable": true,
+          "locationInModule": {
+            "filename": "lib/gateway/gateway.ts",
+            "line": 412
+          },
+          "name": "oauthScopes",
+          "optional": true,
+          "type": {
+            "collection": {
+              "elementtype": {
+                "primitive": "string"
+              },
+              "kind": "array"
+            }
+          }
+        },
         {
           "docs": {
             "custom": {
@@ -13907,6 +13934,23 @@
             }
           }
         },
+        {
+          "docs": {
+            "remarks": "Only available when using the default Cognito authorizer.",
+            "stability": "experimental",
+            "summary": "The OAuth2 token endpoint URL for client credentials flow."
+          },
+          "immutable": true,
+          "locationInModule": {
+            "filename": "lib/gateway/gateway.ts",
+            "line": 406
+          },
+          "name": "tokenEndpointUrl",
+          "optional": true,
+          "type": {
+            "primitive": "string"
+          }
+        },
         {
           "docs": {
             "custom": {
@@ -13927,6 +13971,21 @@
             "primitive": "string"
           }
         },
+        {
+          "docs": {
+            "stability": "experimental",
+            "summary": "The Cognito Resource Server created for the gateway (if using default Cognito authorizer)."
+          },
+          "locationInModule": {
+            "filename": "lib/gateway/gateway.ts",
+            "line": 400
+          },
+          "name": "resourceServer",
+          "optional": true,
+          "type": {
+            "fqn": "aws-cdk-lib.aws_cognito.IUserPoolResourceServer"
+          }
+        },
         {
           "docs": {
             "stability": "experimental",
@@ -13956,6 +14015,21 @@
           "type": {
             "fqn": "aws-cdk-lib.aws_cognito.IUserPoolClient"
           }
+        },
+        {
+          "docs": {
+            "stability": "experimental",
+            "summary": "The Cognito User Pool Domain created for the gateway (if using default Cognito authorizer)."
+          },
+          "locationInModule": {
+            "filename": "lib/gateway/gateway.ts",
+            "line": 395
+          },
+          "name": "userPoolDomain",
+          "optional": true,
+          "type": {
+            "fqn": "aws-cdk-lib.aws_cognito.IUserPoolDomain"
+          }
         }
       ],
       "symbolId": "lib/gateway/gateway:Gateway"
@@ -14222,12 +14296,13 @@
       "methods": [
         {
           "docs": {
+            "remarks": "[disable-awslint:no-grants]",
             "stability": "experimental",
             "summary": "Grants IAM actions to the IAM Principal."
           },
           "locationInModule": {
             "filename": "lib/gateway/gateway-base.ts",
-            "line": 258
+            "line": 260
           },
           "name": "grant",
           "overrides": "@aws-cdk/aws-bedrock-agentcore-alpha.IGateway",
@@ -14261,12 +14336,13 @@
         },
         {
           "docs": {
+            "remarks": "[disable-awslint:no-grants]",
             "stability": "experimental",
             "summary": "Grants permission to invoke this Gateway."
           },
           "locationInModule": {
             "filename": "lib/gateway/gateway-base.ts",
-            "line": 297
+            "line": 305
           },
           "name": "grantInvoke",
           "overrides": "@aws-cdk/aws-bedrock-agentcore-alpha.IGateway",
@@ -14289,12 +14365,13 @@
         },
         {
           "docs": {
+            "remarks": "[disable-awslint:no-grants]",
             "stability": "experimental",
             "summary": "Grants `Create`, `Update`, and `Delete` actions on the Gateway."
           },
           "locationInModule": {
             "filename": "lib/gateway/gateway-base.ts",
-            "line": 288
+            "line": 294
           },
           "name": "grantManage",
           "overrides": "@aws-cdk/aws-bedrock-agentcore-alpha.IGateway",
@@ -14317,12 +14394,13 @@
         },
         {
           "docs": {
+            "remarks": "[disable-awslint:no-grants]",
             "stability": "experimental",
             "summary": "Grants `Get` and `List` actions on the Gateway."
           },
           "locationInModule": {
             "filename": "lib/gateway/gateway-base.ts",
-            "line": 271
+            "line": 275
           },
           "name": "grantRead",
           "overrides": "@aws-cdk/aws-bedrock-agentcore-alpha.IGateway",
@@ -14351,7 +14429,7 @@
           },
           "locationInModule": {
             "filename": "lib/gateway/gateway-base.ts",
-            "line": 314
+            "line": 322
           },
           "name": "metric",
           "overrides": "@aws-cdk/aws-bedrock-agentcore-alpha.IGateway",
@@ -14404,7 +14482,7 @@
           },
           "locationInModule": {
             "filename": "lib/gateway/gateway-base.ts",
-            "line": 368
+            "line": 376
           },
           "name": "metricDuration",
           "overrides": "@aws-cdk/aws-bedrock-agentcore-alpha.IGateway",
@@ -14430,7 +14508,7 @@
           },
           "locationInModule": {
             "filename": "lib/gateway/gateway-base.ts",
-            "line": 327
+            "line": 335
           },
           "name": "metricInvocations",
           "overrides": "@aws-cdk/aws-bedrock-agentcore-alpha.IGateway",
@@ -14457,7 +14535,7 @@
           },
           "locationInModule": {
             "filename": "lib/gateway/gateway-base.ts",
-            "line": 358
+            "line": 366
           },
           "name": "metricLatency",
           "overrides": "@aws-cdk/aws-bedrock-agentcore-alpha.IGateway",
@@ -14483,7 +14561,7 @@
           },
           "locationInModule": {
             "filename": "lib/gateway/gateway-base.ts",
-            "line": 341
+            "line": 349
           },
           "name": "metricSystemErrors",
           "overrides": "@aws-cdk/aws-bedrock-agentcore-alpha.IGateway",
@@ -14510,7 +14588,7 @@
           },
           "locationInModule": {
             "filename": "lib/gateway/gateway-base.ts",
-            "line": 378
+            "line": 386
           },
           "name": "metricTargetExecutionTime",
           "overrides": "@aws-cdk/aws-bedrock-agentcore-alpha.IGateway",
@@ -14536,7 +14614,7 @@
           },
           "locationInModule": {
             "filename": "lib/gateway/gateway-base.ts",
-            "line": 385
+            "line": 393
           },
           "name": "metricTargetType",
           "overrides": "@aws-cdk/aws-bedrock-agentcore-alpha.IGateway",
@@ -14568,7 +14646,7 @@
           },
           "locationInModule": {
             "filename": "lib/gateway/gateway-base.ts",
-            "line": 334
+            "line": 342
           },
           "name": "metricThrottles",
           "overrides": "@aws-cdk/aws-bedrock-agentcore-alpha.IGateway",
@@ -14594,7 +14672,7 @@
           },
           "locationInModule": {
             "filename": "lib/gateway/gateway-base.ts",
-            "line": 348
+            "line": 356
           },
           "name": "metricUserErrors",
           "overrides": "@aws-cdk/aws-bedrock-agentcore-alpha.IGateway",
@@ -15515,13 +15593,13 @@
         },
         {
           "docs": {
-            "remarks": "This method grants the `SynchronizeGatewayTargets` permission, which is primarily\nneeded for MCP Server targets when you need to refresh the tool catalog after the\nMCP server's tools have changed.",
+            "remarks": "This method grants the `SynchronizeGatewayTargets` permission, which is primarily\nneeded for MCP Server targets when you need to refresh the tool catalog after the\nMCP server's tools have changed.\n\n[disable-awslint:no-grants]",
             "stability": "experimental",
             "summary": "Grants permission to synchronize this gateway's targets."
           },
           "locationInModule": {
             "filename": "lib/gateway/targets/target.ts",
-            "line": 501
+            "line": 503
           },
           "name": "grantSync",
           "parameters": [
@@ -15978,12 +16056,13 @@
       "methods": [
         {
           "docs": {
+            "remarks": "[disable-awslint:no-grants]",
             "stability": "experimental",
             "summary": "Grants IAM actions to the IAM Principal."
           },
           "locationInModule": {
             "filename": "lib/gateway/targets/target-base.ts",
-            "line": 160
+            "line": 162
           },
           "name": "grant",
           "overrides": "@aws-cdk/aws-bedrock-agentcore-alpha.IGatewayTarget",
@@ -16017,12 +16096,13 @@
         },
         {
           "docs": {
+            "remarks": "[disable-awslint:no-grants]",
             "stability": "experimental",
             "summary": "Grants `Create`, `Update`, and `Delete` actions on the Gateway Target."
           },
           "locationInModule": {
             "filename": "lib/gateway/targets/target-base.ts",
-            "line": 190
+            "line": 196
           },
           "name": "grantManage",
           "overrides": "@aws-cdk/aws-bedrock-agentcore-alpha.IGatewayTarget",
@@ -16045,12 +16125,13 @@
         },
         {
           "docs": {
+            "remarks": "[disable-awslint:no-grants]",
             "stability": "experimental",
             "summary": "Grants `Get` and `List` actions on the Gateway Target."
           },
           "locationInModule": {
             "filename": "lib/gateway/targets/target-base.ts",
-            "line": 173
+            "line": 177
           },
           "name": "grantRead",
           "overrides": "@aws-cdk/aws-bedrock-agentcore-alpha.IGatewayTarget",
@@ -21272,13 +21353,14 @@
       "methods": [
         {
           "docs": {
+            "remarks": "[disable-awslint:no-grants]",
             "returns": "The Grant object for chaining",
             "stability": "experimental",
             "summary": "Grants the necessary permissions to the role."
           },
           "locationInModule": {
             "filename": "lib/memory/strategies/managed-strategy.ts",
-            "line": 210
+            "line": 213
           },
           "name": "grant",
           "overrides": "@aws-cdk/aws-bedrock-agentcore-alpha.IMemoryStrategy",
@@ -22053,7 +22135,7 @@
         },
         "locationInModule": {
           "filename": "lib/memory/memory.ts",
-          "line": 652
+          "line": 673
         },
         "parameters": [
           {
@@ -22079,7 +22161,7 @@
       "kind": "class",
       "locationInModule": {
         "filename": "lib/memory/memory.ts",
-        "line": 540
+        "line": 561
       },
       "methods": [
         {
@@ -22090,7 +22172,7 @@
           },
           "locationInModule": {
             "filename": "lib/memory/memory.ts",
-            "line": 556
+            "line": 577
           },
           "name": "fromMemoryAttributes",
           "parameters": [
@@ -22137,7 +22219,7 @@
           },
           "locationInModule": {
             "filename": "lib/memory/memory.ts",
-            "line": 734
+            "line": 755
           },
           "name": "addMemoryStrategy",
           "parameters": [
@@ -22161,7 +22243,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "lib/memory/memory.ts",
-            "line": 543
+            "line": 564
           },
           "name": "PROPERTY_INJECTION_ID",
           "static": true,
@@ -22177,7 +22259,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "lib/memory/memory.ts",
-            "line": 634
+            "line": 655
           },
           "name": "grantPrincipal",
           "overrides": "@aws-cdk/aws-bedrock-agentcore-alpha.MemoryBase",
@@ -22196,7 +22278,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "lib/memory/memory.ts",
-            "line": 585
+            "line": 606
           },
           "name": "memoryArn",
           "overrides": "@aws-cdk/aws-bedrock-agentcore-alpha.MemoryBase",
@@ -22215,7 +22297,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "lib/memory/memory.ts",
-            "line": 595
+            "line": 616
           },
           "name": "memoryId",
           "overrides": "@aws-cdk/aws-bedrock-agentcore-alpha.MemoryBase",
@@ -22234,7 +22316,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "lib/memory/memory.ts",
-            "line": 590
+            "line": 611
           },
           "name": "memoryName",
           "type": {
@@ -22252,7 +22334,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "lib/memory/memory.ts",
-            "line": 643
+            "line": 664
           },
           "name": "memoryStrategies",
           "type": {
@@ -22272,7 +22354,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "lib/memory/memory.ts",
-            "line": 620
+            "line": 641
           },
           "name": "createdAt",
           "optional": true,
@@ -22289,7 +22371,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "lib/memory/memory.ts",
-            "line": 608
+            "line": 629
           },
           "name": "description",
           "optional": true,
@@ -22305,7 +22387,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "lib/memory/memory.ts",
-            "line": 612
+            "line": 633
           },
           "name": "executionRole",
           "optional": true,
@@ -22322,7 +22404,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "lib/memory/memory.ts",
-            "line": 599
+            "line": 620
           },
           "name": "expirationDuration",
           "optional": true,
@@ -22341,7 +22423,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "lib/memory/memory.ts",
-            "line": 604
+            "line": 625
           },
           "name": "failureReason",
           "optional": true,
@@ -22357,7 +22439,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "lib/memory/memory.ts",
-            "line": 638
+            "line": 659
           },
           "name": "kmsKey",
           "optional": true,
@@ -22374,7 +22456,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "lib/memory/memory.ts",
-            "line": 616
+            "line": 637
           },
           "name": "status",
           "optional": true,
@@ -22392,7 +22474,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "lib/memory/memory.ts",
-            "line": 630
+            "line": 651
           },
           "name": "tags",
           "optional": true,
@@ -22413,7 +22495,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "lib/memory/memory.ts",
-            "line": 624
+            "line": 645
           },
           "name": "updatedAt",
           "optional": true,
@@ -22440,7 +22522,7 @@
       "kind": "interface",
       "locationInModule": {
         "filename": "lib/memory/memory.ts",
-        "line": 496
+        "line": 517
       },
       "name": "MemoryAttributes",
       "properties": [
@@ -22456,7 +22538,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "lib/memory/memory.ts",
-            "line": 501
+            "line": 522
           },
           "name": "memoryArn",
           "type": {
@@ -22475,7 +22557,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "lib/memory/memory.ts",
-            "line": 506
+            "line": 527
           },
           "name": "roleArn",
           "type": {
@@ -22492,7 +22574,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "lib/memory/memory.ts",
-            "line": 526
+            "line": 547
           },
           "name": "createdAt",
           "optional": true,
@@ -22510,7 +22592,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "lib/memory/memory.ts",
-            "line": 516
+            "line": 537
           },
           "name": "kmsKeyArn",
           "optional": true,
@@ -22528,7 +22610,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "lib/memory/memory.ts",
-            "line": 521
+            "line": 542
           },
           "name": "status",
           "optional": true,
@@ -22546,7 +22628,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "lib/memory/memory.ts",
-            "line": 511
+            "line": 532
           },
           "name": "updatedAt",
           "optional": true,
@@ -22601,13 +22683,14 @@
       "methods": [
         {
           "docs": {
+            "remarks": "[disable-awslint:no-grants]",
             "returns": "An IAM Grant object representing the granted permissions",
             "stability": "experimental",
             "summary": "Grants IAM actions to the IAM Principal."
           },
           "locationInModule": {
             "filename": "lib/memory/memory.ts",
-            "line": 212
+            "line": 215
           },
           "name": "grant",
           "overrides": "@aws-cdk/aws-bedrock-agentcore-alpha.IMemory",
@@ -22642,13 +22725,14 @@
         {
           "docs": {
             "default": "- Default grant configuration:\n- actions: ['bedrock-agentcore:CreateMemory',\n'bedrock-agentcore:GetMemory',\n'bedrock-agentcore:DeleteMemory',\n'bedrock-agentcore:UpdateMemory'] on this.memoryArn",
+            "remarks": "[disable-awslint:no-grants]",
             "returns": "An IAM Grant object representing the granted permissions",
             "stability": "experimental",
             "summary": "Grant the given principal identity permissions to manage the control plane of this memory."
           },
           "locationInModule": {
             "filename": "lib/memory/memory.ts",
-            "line": 323
+            "line": 342
           },
           "name": "grantAdmin",
           "overrides": "@aws-cdk/aws-bedrock-agentcore-alpha.IMemory",
@@ -22672,14 +22756,14 @@
         {
           "docs": {
             "default": "- Default grant configuration:\n- actions: ['bedrock-agentcore:DeleteEvent',\n'bedrock-agentcore:DeleteMemoryRecord'] on this.memoryArn",
-            "remarks": "Both Short-Term Memory (STM) and Long-Term Memory (LTM).",
+            "remarks": "Both Short-Term Memory (STM) and Long-Term Memory (LTM).\n\n[disable-awslint:no-grants]",
             "returns": "An IAM Grant object representing the granted permissions",
             "stability": "experimental",
             "summary": "Grant the given principal identity permissions to delete content on this memory."
           },
           "locationInModule": {
             "filename": "lib/memory/memory.ts",
-            "line": 287
+            "line": 300
           },
           "name": "grantDelete",
           "overrides": "@aws-cdk/aws-bedrock-agentcore-alpha.IMemory",
@@ -22703,13 +22787,14 @@
         {
           "docs": {
             "default": "- Default grant configuration:\n- actions: ['bedrock-agentcore:DeleteMemoryRecord'] on this.memoryArn",
+            "remarks": "[disable-awslint:no-grants]",
             "returns": "An IAM Grant object representing the granted permissions",
             "stability": "experimental",
             "summary": "Grant the given principal identity permissions to delete Long-Term Memory (LTM) content on this memory."
           },
           "locationInModule": {
             "filename": "lib/memory/memory.ts",
-            "line": 309
+            "line": 326
           },
           "name": "grantDeleteLongTermMemory",
           "overrides": "@aws-cdk/aws-bedrock-agentcore-alpha.IMemory",
@@ -22733,13 +22818,14 @@
         {
           "docs": {
             "default": "- Default grant configuration:\n- actions: ['bedrock-agentcore:DeleteEvent'] on this.memoryArn",
+            "remarks": "[disable-awslint:no-grants]",
             "returns": "An IAM Grant object representing the granted permissions",
             "stability": "experimental",
             "summary": "Grant the given principal identity permissions to delete Short-Term Memory (STM) content on this memory."
           },
           "locationInModule": {
             "filename": "lib/memory/memory.ts",
-            "line": 298
+            "line": 313
           },
           "name": "grantDeleteShortTermMemory",
           "overrides": "@aws-cdk/aws-bedrock-agentcore-alpha.IMemory",
@@ -22763,13 +22849,14 @@
         {
           "docs": {
             "default": "- Default grant configuration:\n- actions: ['bedrock-agentcore:CreateEvent',\n'bedrock-agentcore:GetEvent',\n'bedrock-agentcore:DeleteEvent',\n'bedrock-agentcore:GetMemoryRecord',\n'bedrock-agentcore:RetrieveMemoryRecords',\n'bedrock-agentcore:ListMemoryRecords',\n'bedrock-agentcore:ListActors',\n'bedrock-agentcore:ListSessions',\n'bedrock-agentcore:CreateMemory',\n'bedrock-agentcore:GetMemory',\n'bedrock-agentcore:DeleteMemory',\n'bedrock-agentcore:UpdateMemory'] on this.memoryArn",
+            "remarks": "[disable-awslint:no-grants]",
             "returns": "An IAM Grant object representing the granted permissions",
             "stability": "experimental",
             "summary": "Grant the given principal identity permissions to do every action on this memory."
           },
           "locationInModule": {
             "filename": "lib/memory/memory.ts",
-            "line": 345
+            "line": 366
           },
           "name": "grantFullAccess",
           "overrides": "@aws-cdk/aws-bedrock-agentcore-alpha.IMemory",
@@ -22793,14 +22880,14 @@
         {
           "docs": {
             "default": "- Default grant configuration:\n- actions: ['bedrock-agentcore:GetMemoryRecord',\n'bedrock-agentcore:RetrieveMemoryRecords',\n'bedrock-agentcore:ListMemoryRecords',\n'bedrock-agentcore:ListActors',\n'bedrock-agentcore:ListSessions] on this.memoryArn",
-            "remarks": "Both Short-Term Memory (STM) and Long-Term Memory (LTM).",
+            "remarks": "Both Short-Term Memory (STM) and Long-Term Memory (LTM).\n\n[disable-awslint:no-grants]",
             "returns": "An IAM Grant object representing the granted permissions",
             "stability": "experimental",
             "summary": "Grant the given principal identity permissions to read the contents of this memory."
           },
           "locationInModule": {
             "filename": "lib/memory/memory.ts",
-            "line": 244
+            "line": 251
           },
           "name": "grantRead",
           "overrides": "@aws-cdk/aws-bedrock-agentcore-alpha.IMemory",
@@ -22824,13 +22911,14 @@
         {
           "docs": {
             "default": "- Default grant configuration:\n- actions: ['bedrock-agentcore:GetMemoryRecord',\n'bedrock-agentcore:RetrieveMemoryRecords',\n'bedrock-agentcore:ListMemoryRecords',\n'bedrock-agentcore:ListActors',\n'bedrock-agentcore:ListSessions',] on this.memoryArn",
+            "remarks": "[disable-awslint:no-grants]",
             "returns": "An IAM Grant object representing the granted permissions",
             "stability": "experimental",
             "summary": "Grant the given principal identity permissions to read the Long-Term Memory (LTM) contents of this memory."
           },
           "locationInModule": {
             "filename": "lib/memory/memory.ts",
-            "line": 273
+            "line": 284
           },
           "name": "grantReadLongTermMemory",
           "overrides": "@aws-cdk/aws-bedrock-agentcore-alpha.IMemory",
@@ -22854,13 +22942,14 @@
         {
           "docs": {
             "default": "- Default grant configuration:\n- actions: ['bedrock-agentcore:GetEvent',\n'bedrock-agentcore:ListEvents',\n'bedrock-agentcore:ListActors',\n'bedrock-agentcore:ListSessions',] on this.memoryArn",
+            "remarks": "[disable-awslint:no-grants]",
             "returns": "An IAM Grant object representing the granted permissions",
             "stability": "experimental",
             "summary": "Grant the given principal identity permissions to read the Short-Term Memory (STM) contents of this memory."
           },
           "locationInModule": {
             "filename": "lib/memory/memory.ts",
-            "line": 258
+            "line": 267
           },
           "name": "grantReadShortTermMemory",
           "overrides": "@aws-cdk/aws-bedrock-agentcore-alpha.IMemory",
@@ -22884,13 +22973,14 @@
         {
           "docs": {
             "default": "- Default grant configuration:\n- actions: ['bedrock-agentcore:CreateEvent'] on this.memoryArn",
+            "remarks": "[disable-awslint:no-grants]",
             "returns": "An IAM Grant object representing the granted permissions",
             "stability": "experimental",
             "summary": "Grant the given principal identity permissions to write content to short-term memory."
           },
           "locationInModule": {
             "filename": "lib/memory/memory.ts",
-            "line": 228
+            "line": 233
           },
           "name": "grantWrite",
           "overrides": "@aws-cdk/aws-bedrock-agentcore-alpha.IMemory",
@@ -22919,7 +23009,7 @@
           },
           "locationInModule": {
             "filename": "lib/memory/memory.ts",
-            "line": 358
+            "line": 379
           },
           "name": "metric",
           "overrides": "@aws-cdk/aws-bedrock-agentcore-alpha.IMemory",
@@ -22962,7 +23052,7 @@
           },
           "locationInModule": {
             "filename": "lib/memory/memory.ts",
-            "line": 403
+            "line": 424
           },
           "name": "metricErrorsForApiOperation",
           "overrides": "@aws-cdk/aws-bedrock-agentcore-alpha.IMemory",
@@ -22994,7 +23084,7 @@
           },
           "locationInModule": {
             "filename": "lib/memory/memory.ts",
-            "line": 409
+            "line": 430
           },
           "name": "metricEventCreationCount",
           "overrides": "@aws-cdk/aws-bedrock-agentcore-alpha.IMemory",
@@ -23020,7 +23110,7 @@
           },
           "locationInModule": {
             "filename": "lib/memory/memory.ts",
-            "line": 370
+            "line": 391
           },
           "name": "metricForApiOperation",
           "overrides": "@aws-cdk/aws-bedrock-agentcore-alpha.IMemory",
@@ -23058,7 +23148,7 @@
           },
           "locationInModule": {
             "filename": "lib/memory/memory.ts",
-            "line": 394
+            "line": 415
           },
           "name": "metricInvocationsForApiOperation",
           "overrides": "@aws-cdk/aws-bedrock-agentcore-alpha.IMemory",
@@ -23091,7 +23181,7 @@
           },
           "locationInModule": {
             "filename": "lib/memory/memory.ts",
-            "line": 387
+            "line": 408
           },
           "name": "metricLatencyForApiOperation",
           "overrides": "@aws-cdk/aws-bedrock-agentcore-alpha.IMemory",
@@ -23123,7 +23213,7 @@
           },
           "locationInModule": {
             "filename": "lib/memory/memory.ts",
-            "line": 416
+            "line": 437
           },
           "name": "metricMemoryRecordCreationCount",
           "parameters": [
@@ -23303,7 +23393,7 @@
       "kind": "interface",
       "locationInModule": {
         "filename": "lib/memory/memory.ts",
-        "line": 441
+        "line": 462
       },
       "name": "MemoryProps",
       "properties": [
@@ -23316,7 +23406,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "lib/memory/memory.ts",
-            "line": 448
+            "line": 469
           },
           "name": "memoryName",
           "type": {
@@ -23333,7 +23423,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "lib/memory/memory.ts",
-            "line": 462
+            "line": 483
           },
           "name": "description",
           "optional": true,
@@ -23351,7 +23441,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "lib/memory/memory.ts",
-            "line": 480
+            "line": 501
           },
           "name": "executionRole",
           "optional": true,
@@ -23370,7 +23460,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "lib/memory/memory.ts",
-            "line": 455
+            "line": 476
           },
           "name": "expirationDuration",
           "optional": true,
@@ -23388,7 +23478,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "lib/memory/memory.ts",
-            "line": 467
+            "line": 488
           },
           "name": "kmsKey",
           "optional": true,
@@ -23406,7 +23496,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "lib/memory/memory.ts",
-            "line": 473
+            "line": 494
           },
           "name": "memoryStrategies",
           "optional": true,
@@ -23429,7 +23519,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "lib/memory/memory.ts",
-            "line": 487
+            "line": 508
           },
           "name": "tags",
           "optional": true,
@@ -24253,7 +24343,7 @@
       "kind": "interface",
       "locationInModule": {
         "filename": "lib/tools/browser.ts",
-        "line": 483
+        "line": 490
       },
       "name": "RecordingConfig",
       "properties": [
@@ -24267,7 +24357,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "lib/tools/browser.ts",
-            "line": 488
+            "line": 495
           },
           "name": "enabled",
           "optional": true,
@@ -24285,7 +24375,7 @@
           "immutable": true,
           "locationInModule": {
             "filename": "lib/tools/browser.ts",
-            "line": 494
+            "line": 501
           },
           "name": "s3Location",
           "optional": true,
@@ -24973,7 +25063,7 @@
           },
           "locationInModule": {
             "filename": "lib/runtime/runtime-base.ts",
-            "line": 247
+            "line": 249
           },
           "name": "addToRolePolicy",
           "overrides": "@aws-cdk/aws-bedrock-agentcore-alpha.IBedrockAgentRuntime",
@@ -24996,13 +25086,14 @@
         },
         {
           "docs": {
+            "remarks": "[disable-awslint:no-grants]",
             "returns": "The Grant object for chaining",
             "stability": "experimental",
             "summary": "Grant the runtime specific actions on AWS resources."
           },
           "locationInModule": {
             "filename": "lib/runtime/runtime-base.ts",
-            "line": 233
+            "line": 235
           },
           "name": "grant",
           "overrides": "@aws-cdk/aws-bedrock-agentcore-alpha.IBedrockAgentRuntime",
@@ -25044,12 +25135,13 @@
         },
         {
           "docs": {
+            "remarks": "[disable-awslint:no-grants]",
             "stability": "experimental",
             "summary": "Permits an IAM principal to invoke this runtime both directly and on behalf of users Grants both bedrock-agentcore:InvokeAgentRuntime and bedrock-agentcore:InvokeAgentRuntimeForUser permissions."
           },
           "locationInModule": {
             "filename": "lib/runtime/runtime-base.ts",
-            "line": 284
+            "line": 295
           },
           "name": "grantInvoke",
           "overrides": "@aws-cdk/aws-bedrock-agentcore-alpha.IBedrockAgentRuntime",
@@ -25072,12 +25164,13 @@
         },
         {
           "docs": {
+            "remarks": "[disable-awslint:no-grants]",
             "stability": "experimental",
             "summary": "Permits an IAM principal to invoke this runtime Grants the bedrock-agentcore:InvokeAgentRuntime permission."
           },
           "locationInModule": {
             "filename": "lib/runtime/runtime-base.ts",
-            "line": 257
+            "line": 262
           },
           "name": "grantInvokeRuntime",
           "overrides": "@aws-cdk/aws-bedrock-agentcore-alpha.IBedrockAgentRuntime",
@@ -25100,12 +25193,13 @@
         },
         {
           "docs": {
+            "remarks": "[disable-awslint:no-grants]",
             "stability": "experimental",
             "summary": "Permits an IAM principal to invoke this runtime on behalf of a user Grants the bedrock-agentcore:InvokeAgentRuntimeForUser permission Required when using the X-Amzn-Bedrock-AgentCore-Runtime-User-Id header."
           },
           "locationInModule": {
             "filename": "lib/runtime/runtime-base.ts",
-            "line": 271
+            "line": 279
           },
           "name": "grantInvokeRuntimeForUser",
           "overrides": "@aws-cdk/aws-bedrock-agentcore-alpha.IBedrockAgentRuntime",
@@ -25134,7 +25228,7 @@
           },
           "locationInModule": {
             "filename": "lib/runtime/runtime-base.ts",
-            "line": 301
+            "line": 312
           },
           "name": "metric",
           "overrides": "@aws-cdk/aws-bedrock-agentcore-alpha.IBedrockAgentRuntime",
@@ -25177,7 +25271,7 @@
           },
           "locationInModule": {
             "filename": "lib/runtime/runtime-base.ts",
-            "line": 314
+            "line": 325
           },
           "name": "metricInvocations",
           "overrides": "@aws-cdk/aws-bedrock-agentcore-alpha.IBedrockAgentRuntime",
@@ -25203,7 +25297,7 @@
           },
           "locationInModule": {
             "filename": "lib/runtime/runtime-base.ts",
-            "line": 321
+            "line": 332
           },
           "name": "metricInvocationsAggregated",
           "overrides": "@aws-cdk/aws-bedrock-agentcore-alpha.IBedrockAgentRuntime",
@@ -25230,7 +25324,7 @@
           },
           "locationInModule": {
             "filename": "lib/runtime/runtime-base.ts",
-            "line": 352
+            "line": 363
           },
           "name": "metricLatency",
           "overrides": "@aws-cdk/aws-bedrock-agentcore-alpha.IBedrockAgentRuntime",
@@ -25256,7 +25350,7 @@
           },
           "locationInModule": {
             "filename": "lib/runtime/runtime-base.ts",
-            "line": 366
+            "line": 377
           },
           "name": "metricSessionCount",
           "overrides": "@aws-cdk/aws-bedrock-agentcore-alpha.IBedrockAgentRuntime",
@@ -25282,7 +25376,7 @@
           },
           "locationInModule": {
             "filename": "lib/runtime/runtime-base.ts",
-            "line": 373
+            "line": 384
           },
           "name": "metricSessionsAggregated",
           "overrides": "@aws-cdk/aws-bedrock-agentcore-alpha.IBedrockAgentRuntime",
@@ -25308,7 +25402,7 @@
           },
           "locationInModule": {
             "filename": "lib/runtime/runtime-base.ts",
-            "line": 335
+            "line": 346
           },
           "name": "metricSystemErrors",
           "overrides": "@aws-cdk/aws-bedrock-agentcore-alpha.IBedrockAgentRuntime",
@@ -25334,7 +25428,7 @@
           },
           "locationInModule": {
             "filename": "lib/runtime/runtime-base.ts",
-            "line": 328
+            "line": 339
           },
           "name": "metricThrottles",
           "overrides": "@aws-cdk/aws-bedrock-agentcore-alpha.IBedrockAgentRuntime",
@@ -25360,7 +25454,7 @@
           },
           "locationInModule": {
             "filename": "lib/runtime/runtime-base.ts",
-            "line": 359
+            "line": 370
           },
           "name": "metricTotalErrors",
           "overrides": "@aws-cdk/aws-bedrock-agentcore-alpha.IBedrockAgentRuntime",
@@ -25386,7 +25480,7 @@
           },
           "locationInModule": {
             "filename": "lib/runtime/runtime-base.ts",
-            "line": 342
+            "line": 353
           },
           "name": "metricUserErrors",
           "overrides": "@aws-cdk/aws-bedrock-agentcore-alpha.IBedrockAgentRuntime",
@@ -27223,13 +27317,14 @@
       "methods": [
         {
           "docs": {
+            "remarks": "[disable-awslint:no-grants]",
             "returns": "The Grant object for chaining",
             "stability": "experimental",
             "summary": "Grants the necessary permissions to the role."
           },
           "locationInModule": {
             "filename": "lib/memory/strategies/self-managed-strategy.ts",
-            "line": 226
+            "line": 229
           },
           "name": "grant",
           "overrides": "@aws-cdk/aws-bedrock-agentcore-alpha.IMemoryStrategy",
@@ -28178,6 +28273,6 @@
       "symbolId": "lib/network/network-configuration:VpcConfigProps"
     }
   },
-  "version": "2.234.0-alpha.0",
+  "version": "2.235.0-alpha.0",
   "fingerprint": "**********"
 }