AWS Bedrock Agent Codex

This repository provides an example project for deploying an AWS Bedrock Agent using Terraform and GitHub Actions. It also includes a sample Python Lambda function that the agent can invoke to perform custom actions.

Project structure

.
├── lambda/                # Python source code for the Lambda function
│   ├── app.py             # Lambda handler
│   └── requirements.txt   # Python dependencies
├── terraform/             # Terraform configuration to deploy the agent and Lambda
│   ├── main.tf
│   ├── outputs.tf
│   ├── provider.tf
│   └── variables.tf
└── .github/workflows/     # GitHub Actions workflows
    └── terraform.yml      # CI/CD pipeline

Terraform

The Terraform configuration creates an IAM role, deploys the Lambda function and provides a placeholder for future Bedrock Agent resources. Customize terraform/main.tf once the Bedrock Agent Terraform resources become available.

Run the following commands locally to deploy:

terraform -chdir=terraform init
terraform -chdir=terraform plan
terraform -chdir=terraform apply

GitHub Actions workflow

The workflow defined in .github/workflows/terraform.yml automatically packages the Lambda code, initializes Terraform and applies the configuration after a pull request is merged into the main branch.

Lambda function

The sample Lambda function located in lambda/app.py simply logs the incoming event and returns a JSON payload. Extend this file with any custom logic needed by your Bedrock Agent.

Getting started

1. Ensure you have Terraform installed.
2. Configure AWS credentials in your environment (e.g., using aws configure).
3. Package the Lambda function and apply the Terraform configuration.

This setup demonstrates how infrastructure as code and automation can be combined to manage Bedrock Agents on AWS.