cloudwatch_skill

This skill helps you monitor AWS resources with CloudWatch, create alarms, query logs, and build dashboards for proactive observability.
  • 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 cloudwatch

  • alarms-metrics.md11.0 KB
  • SKILL.md10.3 KB

Overview

This skill provides practical AWS CloudWatch monitoring and observability workflows for logs, metrics, alarms, and dashboards. It focuses on creating alarms, querying logs with Insights, configuring metric filters, publishing custom metrics, and building dashboards to troubleshoot and optimize applications. Use it to speed up setup and operational tasks with CLI and boto3 examples.

How this skill works

The skill inspects CloudWatch concepts and common operational patterns, then generates concrete commands and code snippets to implement them (CLI and boto3). It covers creating metric alarms, composite or expression-based alarms, running Logs Insights queries, creating metric filters from log patterns, publishing custom metrics, and assembling dashboards. It also highlights troubleshooting checks and cost-optimization tips to quickly diagnose issues like missing metrics or stuck alarms.

When to use it

  • Setting up monitoring for EC2, Lambda, RDS, or custom applications
  • Creating alarms for CPU, error rate, latency, or composite conditions
  • Querying logs with Logs Insights to find recent errors or patterns
  • Extracting metrics from logs via metric filters and publishing custom metrics
  • Building dashboards to correlate metrics and recent log activity

Best practices

  • Use dimensions sparingly to avoid metric explosion and high cardinality
  • Aggregate and batch custom metrics to reduce API calls and cost
  • Prefer structured JSON logs for reliable parsing and metric extraction
  • Set appropriate evaluation periods and use anomaly detection to avoid flapping
  • Apply retention policies and remove unused dashboards to control costs

Example use cases

  • Create an EC2 CPU utilization alarm with SNS notifications and OK actions
  • Build a Lambda error-rate alarm using metric math (errors/invocations*100)
  • Run a Logs Insights query to list recent ERROR messages from a Lambda
  • Add a metric filter that increments ErrorCount when logs contain 'ERROR'
  • Publish OrdersProcessed as a custom metric with Environment and OrderType dimensions

FAQ

Check that the metric is actually publishing, verify namespace and dimensions, and increase evaluation periods or ensure detailed monitoring is enabled.

How do I reduce CloudWatch costs?

Lower log retention for non-critical groups, delete unused dashboards, avoid unnecessary high-resolution metrics, and batch metric submissions.

Built by
VeilStrat
AI signals for GTM teams
© 2026 VeilStrat. All rights reserved.All systems operational
cloudwatch skill by itsmostafa/aws-agent-skills | VeilStrat