| name | powertools-aws |
| description | ALWAYS use this skill when writing AWS Lambda functions in Python, TypeScript, Java, or .NET. Provides serverless best practices for structured logging, distributed tracing (X-Ray), CloudWatch metrics, idempotency, batch processing, and event handling using Powertools for AWS Lambda. Essential for any Lambda development. |
| license | MIT |
| metadata | {"author":"aws-powertools","repository":"https://github.com/aws-powertools","runtimes":"python, typescript, java, dotnet"} |
Powertools for AWS Lambda
A developer toolkit to implement Serverless best practices and increase developer velocity across Python, TypeScript, Java, and .NET Lambda runtimes.
Invoke This Skill When
- Writing or modifying AWS Lambda functions
- User mentions "serverless" with Python, TypeScript, Java, or .NET
- Implementing structured logging in Lambda
- Adding distributed tracing (AWS X-Ray)
- Creating CloudWatch custom metrics or Datadog metrics
- Handling SQS, Kinesis, or DynamoDB Streams batch processing
- Implementing idempotency patterns
- Building REST or GraphQL APIs with Lambda
- Building AppSync resolvers or event handlers
- Creating Amazon Bedrock Agents with Lambda
- Processing Kafka events (MSK, self-managed)
- User mentions Powertools, aws-lambda-powertools, or @aws-lambda-powertools
- Working with SAM, CDK, or Serverless Framework templates
- Needing to parse/validate Lambda event payloads
- Implementing feature flags or data masking
Documentation Sources (llms.txt)
CRITICAL: Before implementing any Powertools feature, fetch the llms.txt for the target runtime. These files contain the authoritative, up-to-date documentation index.
Note: .NET does not use /latest/ in its URL path.
How to Use Documentation
- Identify the runtime from file extensions or user request
- Fetch the llms.txt for that runtime
- Find the relevant section (e.g., Logger, Metrics, Batch)
- Fetch the specific doc page (URLs in llms.txt point to
.md files)
- Apply the patterns from the documentation
Runtime Detection
| File Pattern | Runtime |
|---|
*.py with def handler(event, context) | Python |
*.ts or *.js with export const handler | TypeScript |
*.java with handleRequest method | Java |
*.cs with FunctionHandler | .NET |
template.yaml / samconfig.toml | Check Runtime property |
serverless.yml | Check provider.runtime |
cdk.json / *.cdk.ts | Check runtime in Lambda construct |
Core Utilities Overview
All runtimes provide these core utilities:
| Utility | Purpose | When to Use |
|---|
| Logger | Structured JSON logging with Lambda context | Every Lambda - always add logging |
| Tracer | AWS X-Ray distributed tracing | Debugging, performance analysis, service maps |
| Metrics | CloudWatch Embedded Metric Format (EMF) | Monitoring, alerting, dashboards |
Additional Utilities (varies by runtime)
| Utility | Python | TypeScript | Java | .NET | Purpose |
|---|
| Idempotency | Y | Y | Y | Y | Prevent duplicate processing |
| Batch Processing | Y | Y | Y | Y | Handle partial failures (SQS/Kinesis/DynamoDB) |
| Parameters | Y | Y | Y | Y | SSM/Secrets Manager with caching |
| Kafka Consumer | Y | Y | Y | Y | Kafka event deserialization (Avro, Protobuf, JSON Schema) |
| JMESPath Functions | Y | Y | - | Y | Extract/deserialize nested JSON payloads |
| Parser/Validation | Y | Y | Y | - | Event validation and parsing (Pydantic/Standard Schema) |
| Event Handler: REST API | Y | Y | - | - | API Gateway, ALB, Lambda Function URL |
| Event Handler: GraphQL | Y | Y | - | - | AppSync GraphQL resolvers |
| Event Handler: AppSync Events | Y | Y | - | Y | AppSync real-time event handlers |
| Event Handler: Bedrock Agents | Y | Y | - | Y | Amazon Bedrock Agent function resolvers |
| Feature Flags | Y | - | - | - | Runtime feature toggling with AppConfig |
| Streaming | Y | - | - | - | Process large S3 objects/datasets |
| Data Masking | Y | - | - | - | Encrypt/mask sensitive data fields |
| Event Source Data Classes | Y | - | - | - | Typed classes for Lambda event sources |
| Typing | Y | - | - | - | Static typing for Lambda context |
| Middleware Factory | Y | - | - | - | Create custom middleware decorators |
| Datadog Metrics | Y | - | - | - | Alternative metrics provider for Datadog |
| Custom Resources | - | - | Y | - | CloudFormation custom resource handlers |
| Large Messages | - | - | Y | - | Handle messages >256KB via S3 |
| Serialization | - | - | Y | - | JSON serialization utilities |
