Deploy to AWS Lambda

Serverless deployment on AWS using SAM (Serverless Application Model).

Prerequisites

Step-by-step

1

Install AWS CLI and SAM CLI

# macOS
brew install awscli aws-sam-cli

# Linux
pip install aws-sam-cli

# Verify installation
aws --version
sam --version
2

Configure AWS credentials

aws configure
# Enter your:
#   AWS Access Key ID
#   AWS Secret Access Key
#   Default region (e.g., us-east-1)
#   Default output format (json)

Get your access keys from the AWS IAM Console.

3

Create your project with AWS Lambda target

npx agentvoy create my-project --build-mode app --deploy-target aws-lambda --yes
4

Configure your .env

cd my-project-app
cp .env.example .env
# Add your API key to .env
5

Deploy

agentvoy deploy --target aws-lambda

AgentVoy runs sam build and sam deploy --guided for you.

6

Access your API

# Get the API Gateway URL
sam list stack-outputs --stack-name my-project

# View logs
sam logs -n AgentFunction --stack-name my-project --tail

Generated files

+ Dockerfile
+ .dockerignore
+ deploy/template.yaml (SAM template)
+ deploy/lambda_handler.py

Important notes

AWS Lambda has a 15-minute maximum execution timeout. Long-running agent tasks may need to be broken into smaller steps.

The DevTools WebSocket endpoint (/ws/trace) is not available on Lambda since it does not support persistent connections. Use CloudWatch Logs instead.