Strands Agent Example

An example app for building personal agentic AI backends using AWS Lambda, API Gateway, and Strands Agents.
This project demonstrates a hybrid TypeScript (AWS CDK) and Python (with AWS Powertools) workflow, with poetry dependency management and Lambda Layer separation.

🛠️ Project Structure

.
├── app/                # Python Lambda source (Powertools, Strands Agent logic)
│   ├── agent.py
│   └── main.py
├── lambda/layer/       # Lambda Layer for Python dependencies (built via Poetry/Makefile)
├── lib/                # CDK TypeScript stack (API Gateway, Lambda, Layer)
├── bin/                # CDK entrypoint
├── test/               # CDK/infra tests
├── pyproject.toml      # Poetry project config for lambda dependencies
├── package.json        # Node/TypeScript project config for CDK
├── Makefile            # Build, lint, and layer automation
└── ...

🚀 Quickstart

1. Initialize the Project (Recommended for new users)

make

• Installs Node/TypeScript and Python dependencies (via Poetry)
• Installs the pre-commit hook for lint and test

2. Build the Lambda Layer

make layer

• Exports Python dependencies from Poetry and installs them into lambda/layer/python for Lambda Layer packaging.

3. Deploy to AWS

npx cdk deploy

• Deploys the stack: Lambda Layer, Lambda function, and API Gateway.

🧹 Pre-commit Hook (via Husky)

This repo uses Husky to run make lint-check and npm test before every commit. The hook is installed automatically with make or make setup.

To (re)install the hook manually:

make setup

To customize the hook, edit .husky/pre-commit.

🔁 API Usage

After deployment, your API Gateway endpoint will look like:

POST https://<api-id>.execute-api.<region>.amazonaws.com/v1/invoke

Example request:

curl -X POST \
  -H "Content-Type: application/json" \
  -d '{"prompt": "Hello, agent!"}' \
  https://<api-id>.execute-api.<region>.amazonaws.com/v1/invoke

Expected response:

{"message": "<agent response>"}

💸 Cost Estimate

Running this stack is very affordable for most personal and prototype use cases, but here are the main cost factors:

• AWS Lambda:

  • Free tier: 1M requests and 400,000 GB-seconds compute per month (Lambda pricing).
  • After free tier: ~$0.20 per 1M requests, plus compute time (depends on memory and duration).

• API Gateway:

  • Free tier: 1M HTTP API calls per month (API Gateway pricing).
  • After free tier: ~$1.00 per 1M requests.

• Lambda Layer storage:

  • No extra charge for using Lambda Layers, but you pay for total Lambda storage if you have many versions (Lambda storage pricing).

• Amazon Bedrock Model Inference:

  • Each API call invokes an Amazon Bedrock model (by default currently in strands, Sonnet 3.7).
  • Pricing depends on the model and tokens processed (Bedrock pricing).
  • As of July 2025, Sonnet 3.7 is ~$0.0030 per 1,000 input tokens and ~$0.0150 per 1,000 output tokens (check the link for current rates).

Example:

If you run 100,000 invocations per month, each <1s and <512MB, and each prompt/response averages 1,000 tokens in and 1,000 tokens out:

• Lambda and API Gateway: Likely free (within free tier) for most new AWS accounts; after free tier, well under $1/month for typical usage.
• Bedrock Sonnet 3.7: ~$0.0030 + $0.0150 = $0.018 per invocation → $1.80 per 100,000 invocations (token usage will vary).

For up-to-date details, always check the official AWS pricing pages linked above.

🪪 License

MIT