Use CreateAgent with an AWS SDK or CLI - Amazon Bedrock

Use CreateAgent with an AWS SDK or CLI

The following code examples show how to use CreateAgent.

Action examples are code excerpts from larger programs and must be run in context. You can see this action in context in the following code example:

JavaScript
SDK for JavaScript (v3)
Note

There's more on GitHub. Find the complete example and learn how to set up and run in the AWS Code Examples Repository.

Create an agent.

// Copyright Amazon.com, Inc. or its affiliates. All Rights Reserved. // SPDX-License-Identifier: Apache-2.0 import { fileURLToPath } from "url"; import { checkForPlaceholders } from "../lib/utils.js"; import { BedrockAgentClient, CreateAgentCommand, } from "@aws-sdk/client-bedrock-agent"; /** * Creates an Amazon Bedrock Agent. * * @param {string} agentName - A name for the agent that you create. * @param {string} foundationModel - The foundation model to be used by the agent you create. * @param {string} agentResourceRoleArn - The ARN of the IAM role with permissions required by the agent. * @param {string} [region='us-east-1'] - The AWS region in use. * @returns {Promise<import("@aws-sdk/client-bedrock-agent").Agent>} An object containing details of the created agent. */ export const createAgent = async ( agentName, foundationModel, agentResourceRoleArn, region = "us-east-1", ) => { const client = new BedrockAgentClient({ region }); const command = new CreateAgentCommand({ agentName, foundationModel, agentResourceRoleArn, }); const response = await client.send(command); return response.agent; }; // Invoke main function if this file was run directly. if (process.argv[1] === fileURLToPath(import.meta.url)) { // Replace the placeholders for agentName and accountId, and roleName with a unique name for the new agent, // the id of your AWS account, and the name of an existing execution role that the agent can use inside your account. // For foundationModel, specify the desired model. Ensure to remove the brackets '[]' before adding your data. // A string (max 100 chars) that can include letters, numbers, dashes '-', and underscores '_'. const agentName = "[your-bedrock-agent-name]"; // Your AWS account id. const accountId = "[123456789012]"; // The name of the agent's execution role. It must be prefixed by `AmazonBedrockExecutionRoleForAgents_`. const roleName = "[AmazonBedrockExecutionRoleForAgents_your-role-name]"; // The ARN for the agent's execution role. // Follow the ARN format: 'arn:aws:iam::account-id:role/role-name' const roleArn = `arn:aws:iam::${accountId}:role/${roleName}`; // Specify the model for the agent. Change if a different model is preferred. const foundationModel = "anthropic.claude-v2"; // Check for unresolved placeholders in agentName and roleArn. checkForPlaceholders([agentName, roleArn]); console.log(`Creating a new agent...`); const agent = await createAgent(agentName, foundationModel, roleArn); console.log(agent); }
  • For API details, see CreateAgent in AWS SDK for JavaScript API Reference.

Python
SDK for Python (Boto3)
Note

There's more on GitHub. Find the complete example and learn how to set up and run in the AWS Code Examples Repository.

Create an agent.

def create_agent(self, agent_name, foundation_model, role_arn, instruction): """ Creates an agent that orchestrates interactions between foundation models, data sources, software applications, user conversations, and APIs to carry out tasks to help customers. :param agent_name: A name for the agent. :param foundation_model: The foundation model to be used for orchestration by the agent. :param role_arn: The ARN of the IAM role with permissions needed by the agent. :param instruction: Instructions that tell the agent what it should do and how it should interact with users. :return: The response from Agents for Bedrock if successful, otherwise raises an exception. """ try: response = self.client.create_agent( agentName=agent_name, foundationModel=foundation_model, agentResourceRoleArn=role_arn, instruction=instruction, ) except ClientError as e: logger.error(f"Error: Couldn't create agent. Here's why: {e}") raise else: return response["agent"]
  • For API details, see CreateAgent in AWS SDK for Python (Boto3) API Reference.

For a complete list of AWS SDK developer guides and code examples, see Using this service with an AWS SDK. This topic also includes information about getting started and details about previous SDK versions.