AWS SAM Bootstrap

Overview

Bootstrap AWS SAM projects end-to-end for both greenfield and migration scenarios. Generate the minimum required artifacts for reliable sam build and sam deploy workflows while following AWS SAM conventions.

When to Use

Use this skill when:

• Configuring AWS SAM for an existing Lambda or CloudFormation-based project
• Creating a new serverless project with AWS SAM from scratch
• Adding template.yaml and samconfig.toml to standardize deployments
• Setting up local testing with sam local invoke and sample event payloads
• Defining environment-aware deploy settings (for example default and prod)

Trigger phrases:

• "Configure AWS SAM for this project"
• "Create a new SAM project"
• "Add template.yaml for SAM"
• "Initialize SAM for existing Lambda function"
• "Generate samconfig.toml"

Instructions

1) Classify the Scenario

Identify one path before writing files:

• New project: no deployable Lambda structure exists yet.
• Existing project migration: Lambda code and/or CloudFormation resources already exist.

2) Choose Runtime and Package Type

Select runtime and package mode first, then keep all generated files consistent.

Supported runtime families:

• Python: python3.10 to python3.14 (avoid 3.8/3.9, deprecated)
• Node.js: nodejs20.x, nodejs22.x, nodejs24.x (avoid 18.x, deprecated)
• Java: java8.al2, java11, java17, java21, java25
• Go: provided.al2, provided.al2023
• .NET: dotnet8, dotnet9, dotnet10 (avoid dotnet6, deprecated)
• Ruby: ruby3.2, ruby3.3, ruby3.4

Package type:

• Zip: default for most functions
• Image: choose when container packaging or native dependencies are needed

3) Bootstrap New Projects

For new projects:

1. Run sam init with explicit runtime, architecture, and package type.
2. Keep the generated structure minimal and production-oriented.
3. Add/adjust samconfig.toml with environment-specific deploy settings.
4. Add events/ payloads for local test commands.

Recommended command sequence:

sam init
sam build
sam local invoke <LogicalFunctionId> -e events/event.json
sam deploy --guided

4) Bootstrap Existing Projects

For existing projects:

1. Inspect current Lambda handlers, runtime, and dependency layout.
2. Create template.yaml with Transform: AWS::Serverless-2016-10-31.
3. Map existing resources into AWS::Serverless::Function and related SAM resources.
4. Create samconfig.toml with deploy defaults and environment overrides.
5. Add events/ payload samples for local invocation.
6. Validate with sam validate and sam build before deploy.

Migration sequence:

sam validate
sam build
sam local invoke <LogicalFunctionId> -e events/event.json
sam deploy --guided

5) Ensure Required Artifacts Exist

Always produce and verify these artifacts:

.
├── template.yaml
├── samconfig.toml
└── events/
    └── event.json

Use reference templates from:

• SAM Template Examples
• Migration Checklist

6) Configure samconfig.toml Correctly

At minimum:

• stack_name
• capabilities (typically CAPABILITY_IAM)
• resolve_s3 = true
• environment separation (default, optional prod)
• optional build acceleration (cached = true, parallel = true)

Baseline example:

version = 0.1

[default.global.parameters]
stack_name = "my-sam-app"

[default.build.parameters]
cached = true
parallel = true

[default.deploy.parameters]
capabilities = "CAPABILITY_IAM"
confirm_changeset = true
resolve_s3 = true

[prod.deploy.parameters]
confirm_changeset = false

7) Standard Command Coverage

Use these commands according to task phase:

• sam init for new projects
• sam build to compile/package artifacts
• sam deploy --guided for first deployment configuration
• sam package when explicit packaging step is required
• sam local invoke for local function execution with event payloads

8) Validate Before Finalizing

Run this verification checklist:

• sam validate succeeds
• sam build succeeds
• template.yaml has correct logical IDs and handlers
• samconfig.toml contains deploy parameters for target environments
• events/event.json matches handler input expectations

Examples

Example A: Create a New SAM App

User request: "Create a new SAM project for a Python Lambda API"

Response approach:

1. Initialize with sam init and Python runtime.
2. Confirm package type (Zip by default).
3. Generate samconfig.toml with default and prod deploy parameters.
4. Add events/event.json.
5. Validate with sam build and sam local invoke.

Input:

Create a new SAM project for a Python Lambda API with a /hello endpoint.

Output:

Generated files:
- template.yaml (AWS::Serverless::Function + API event)
- samconfig.toml (default/prod deploy settings)
- events/event.json

Commands to run:
sam build
sam local invoke HelloFunction -e events/event.json
sam deploy --guided

Example B: Migrate Existing Lambda Project

User request: "Configure AWS SAM for this existing Lambda project"

Response approach:

1. Detect current handler/runtime from the codebase.
2. Create template.yaml with SAM transform and function definition.
3. Generate samconfig.toml including stack_name, capabilities, and resolve_s3.
4. Add local test events.
5. Run validate/build/deploy workflow commands.

Input:

Configure AWS SAM for this existing Lambda project and prepare deployment artifacts.

Output:

Migration result:
- Added template.yaml with Transform AWS::Serverless-2016-10-31
- Added samconfig.toml with deployment parameters
- Added events/event.json for local testing
- Validation workflow prepared:
  sam validate
  sam build
  sam deploy --guided

Best Practices

• Keep template resources explicit and minimal; avoid unrelated infrastructure in migration-first PRs
• Prefer one deployable function flow first, then expand to multiple functions
• Keep samconfig.toml committed for deterministic deployments
• Use environment-specific sections instead of ad-hoc CLI flags
• Define IAM permissions with least privilege
• Add local events that reflect real payloads used in production integrations

Constraints and Warnings

• SAM CLI must be installed locally for command execution
• CAPABILITY_IAM is required when IAM resources are created
• Container image packaging requires Docker availability
• Existing projects may require refactoring handler paths to match SAM conventions
• sam deploy --guided writes local configuration; review before committing