Monitoring, observability, operational procedures, CI/CD lessons learned, and quality gate definitions. Load when assessing production readiness or validating operational excellence.
SkillScore95Maintained
On-demandOn-demand: invoked explicitly — never occupies your agent's context until called.
◆DevOps & SRE pack
Last verified 8/16/2026
What it changes
Passed 21 of 22 test cases with the skill, and 15 of 22 without it.
Without skill68%
With skill95%
Checks it passed that the model alone missed
Specifies tracking cache hit rates.
Assigns next business day review to the Info alert tier.
Specifies performing a post-incident review within 48 hours.
and 7 more in the full report
6 fixed15 passed either way1 still failing0 broke
Pass-rate lift
+27 pp
Tokens
-16%
Agent turns
0%
Response time
-37%
Model gemini-3.6-flashMeasured 8/16/2026v121 of 22 cases passed with skill
Tokens, turns and time describe resource use, not a dollar price.
See it in action
One of 6 test cases the skill fixed. Same model, same prompt, with and without the skill.
1 of 6
Prompt
What non-urgent system events qualify for the Info alert tier, and when should on-call engineers review them?
What was checked, without and with the skill
Checked
Without skill
With skill
Identifies deployment complete as an Info tier condition.
Passed
Passed
Specifies reviewing Info tier items on the next business day.
Missed
Passed
case-19 · gemini-3.6-flash · 8/16/2026 · answers are capped and redacted
What's inside
The instructions your model reads when the skill loads.
Server/container health and resource utilization | Network performance and connectivity | Storage capacity and I/O performance | Security event detection.
Alerting Tiers
Tier
Condition
Response
Page
Service down, data loss risk, security breach
Immediate response
Urgent
Error rate >2x baseline, latency SLA breach
Response within 15 min
Warning
Capacity >80%, error rate trending up
Response within 1 hour
Info
Deployment complete, metric threshold crossed
Review next business day
Operational Procedures
Incident Response
Detect: automated alerting identifies issue
Triage: classify severity, assign responder
Communicate: notify stakeholders per severity level
Resolve: apply fix or rollback
Review: post-incident review within 48 hours
Improve: update runbooks and monitoring based on findings
Operational runbooks for common procedures | Architecture documentation with system diagrams | Deployment procedures and configuration management | Troubleshooting guides for known failure modes.