Perform AI prompt-chaining with Amazon Bedrock - AWS Step Functions
PrerequisitesStep 1: Create the state machineStep 2: Run the state machine

Perform AI prompt-chaining with Amazon Bedrock

This sample project demonstrates how you can integrate with Amazon Bedrock to perform AI prompt-chaining and build high-quality chatbots using Amazon Bedrock. The project chains together some prompts and resolves them in the sequence in which they're provided. Chaining of these prompts augments the ability of the language model being used to deliver a highly-curated response.

This sample project creates the state machine, the supporting AWS resources, and configures the related IAM permissions. Explore this sample project to learn about using Amazon Bedrock optimized service integration with Step Functions state machines, or use it as a starting point for your own projects.

Prerequisites

This sample project uses the Cohere Command large language model (LLM). To successfully run this sample project, you must add access to this LLM from the Amazon Bedrock console. To add the model access, do the following:

  1. Open the Amazon Bedrock console.

  2. On the navigation pane, choose Model access.

  3. Choose Manage model access.

  4. Select the check box next to Cohere.

  5. Choose Request access. The Access status for Cohere model shows as Access granted.

Step 1: Create the state machine

  1. Open the Step Functions console and choose Create state machine.

  2. Find and choose the starter template you want to work with. Choose Next to continue.

  3. Choose Run a demo to create a read-only and ready-to-deploy workflow, or choose Build on it to create an editable state machine definition that you can build on and later deploy.

  4. Choose Use template to continue with your selection.

Next steps depend on your previous choice:

  1. Run a demo – You can review the state machine before you create a read-only project with resources deployed by AWS CloudFormation to your AWS account.

    You can view the state machine definition, and when you are ready, choose Deploy and run to deploy the project and create the resources.

    Deploying can take up to 10 minutes to create resources and permissions. You can use the Stack ID link to monitor progress in AWS CloudFormation.

    After deploy completes, you should see your new state machine in the console.

  2. Build on it – You can review and edit the workflow definition. You might need to set values for placeholders in the sample project before attemping to run your custom workflow.

Note

Standard charges might apply for services deployed to your account.

Step 2: Run the state machine

  1. On the State machines page, choose your sample project.

  2. On the sample project page, choose Start execution.

  3. In the Start execution dialog box, do the following:

    1. (Optional) Enter a custom execution name to override the generated default.

      Non-ASCII names and logging

      Step Functions accepts names for state machines, executions, activities, and labels that contain non-ASCII characters. Because such characters will not work with Amazon CloudWatch, we recommend using only ASCII characters so you can track metrics in CloudWatch.

    2. (Optional) In the Input box, enter input values as JSON. You can skip this step if you are running a demo.

    3. Choose Start execution.

    The Step Functions console will direct you to an Execution Details page where you can choose states in the Graph view to explore related information in the Step details pane.

Congratulations!

You should now have either a running demo or a state machine definition that you can customize.