devops-deployment-safety-check

Validate deployment readiness and monitor canary rollout health using AWS CloudWatch and ELB metrics.

Updated Apr 5, 2026
One-click install
npx skills add https://github.com/schoolofdevops/401 --skill devops-deployment-safety-check-schoolofdevops
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: devops-deployment-safety-check
Source: https://github.com/schoolofdevops/401/tree/main/agents/track-b-finops/skills/devops-deployment-safety-check
Command: npx skills add https://github.com/schoolofdevops/401 --skill devops-deployment-safety-check-schoolofdevops

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Deploying to production without verifying target health, active alarms, and error baselines risks outages, and canary releases need objective pass/fail criteria to decide between full rollout and rollback. ## Core Features & Use Cases - Pre-Deployment Gate: Checks target group health, 5XX error baselines, active CloudWatch alarms, and minimum instance counts before allowing a deploy. - Canary Health Monitoring: Polls 5XX error rates and p95 response times every 2 minutes during a 10-minute canary window and compares them against the pre-deploy baseline. - Gate and Rollback Decisions: Applies explicit numeric thresholds to output PASS, BLOCK, or ROLLBACK decisions with escalation handoffs. - Use Case: A CI/CD pipeline triggers this gate before promoting api-service to production; the skill blocks the deploy because a CloudWatch alarm is in ALARM state, preventing a risky release. ## Quick Start Ask the agent to run the deployment safety gate for api-service in production before the canary release begins.

Frequently Asked Questions about devops-deployment-safety-check

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I run a pre-deployment safety check before deploying to AWS production?

Set DEPLOYMENT_TARGET, DEPLOYMENT_ENV, and AWS_DEFAULT_REGION, then run the gate steps: check target group health, 5XX error baseline, active CloudWatch alarms, and minimum instance count. The deployment is blocked if any alarm is in ALARM state or any target is unhealthy.

How to monitor canary deployment health with CloudWatch metrics?

Poll HTTPCode_Target_5XX_Count and TargetResponseTime p95 every 2 minutes during a minimum 10-minute canary window. Compare each reading against the pre-deploy baseline; rollback triggers if 5XX triples the baseline or p95 exceeds 2000ms for two consecutive windows.

What AWS permissions are needed for a deployment readiness gate?

The gate requires read-only permissions: elasticloadbalancing:DescribeTargetHealth and DescribeTargetGroups, cloudwatch:GetMetricStatistics and DescribeAlarms, and ec2:DescribeInstances. No write permissions are needed since the skill only outputs gate decisions.

Can this skill execute deployments or rollbacks directly?

No. The skill is a gate, not a deployer; it outputs PASS, BLOCK, or ROLLBACK decisions that the CI/CD pipeline executes. Rollback via aws codedeploy stop-deployment requires explicit on-call engineer approval.

When should I not use a deployment safety gate skill?

Do not use it for routine health monitoring without an active deployment, cost investigations, or Kubernetes pod restarts unrelated to a deployment. It is designed only for pre-deploy validation, canary monitoring, and rollback decisions.