sqs_skill
- Python
976
GitHub Stars
2
Bundled Files
2 months ago
Catalog Refreshed
4 months ago
First Indexed
Readme & install
Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.
Installation
Preview and clipboard use veilstrat where the catalogue uses aiagentskills.
npx veilstrat add skill itsmostafa/aws-agent-skills --skill sqs- messaging-patterns.md10.5 KB
- SKILL.md8.8 KB
Overview
This skill provides programmatic control over AWS SQS for building decoupled, resilient architectures. It helps create and configure standard and FIFO queues, manage dead-letter queues, tune visibility timeouts, and integrate queues with Lambda. Use it to automate messaging patterns, batching, and monitoring.
How this skill works
The skill exposes common SQS operations via CLI and boto3 examples so agents can create queues, set attributes, send and receive messages, and manage DLQs. It covers FIFO semantics, message batching, long polling, visibility changes, and event source mappings for Lambda. Troubleshooting commands and monitoring tips are included to detect inflight, delayed, or dead-lettered messages.
When to use it
- Decouple microservices or serverless components to improve scalability and reliability
- Create FIFO queues when strict ordering or exactly-once processing is required
- Configure dead-letter queues to capture and inspect failing messages
- Integrate SQS with Lambda for event-driven processing and batch handling
- Optimize polling and visibility for cost and throughput control
Best practices
- Enable long polling (WaitTimeSeconds=20) to lower API calls and costs
- Always configure a DLQ and set maxReceiveCount (commonly 3–5) for production queues
- Choose FIFO with MessageGroupId and deduplication for order-sensitive workflows
- Set visibility timeout longer than peak processing time and implement idempotent consumers
- Encrypt queues (SSE-SQS or KMS) and use queue policies or VPC endpoints for secure access
Example use cases
- Order processing pipeline: send orders to a FIFO queue with MessageGroupId to preserve per-customer order
- Async task workers: standard queues distribute work across autoscaled consumers with long polling
- Error handling: route repeatedly failing messages to a DLQ for manual or automated inspection
- Lambda-driven ETL: event source mapping batches messages into a Lambda function for transformation
- Backpressure smoothing: buffer bursts from incoming traffic to downstream services
FAQ
Use FIFO when you need strict ordering and exactly-once processing. Use Standard for high throughput where occasional duplicates or out-of-order delivery is acceptable.
How do I prevent duplicate processing?
Use FIFO with deduplication or implement idempotent consumers that track processed message IDs. Also tune visibility timeout and delete messages immediately after successful processing.