Identify the primary drivers of a sudden cloud cost increase and implement safe mitigations.
SkillScore59Maintained
Automatic
◆DevOps & SRE pack
Last verified 8/3/2026
What it changes
Passed 13 of 22 test cases with the skill, and 11 of 22 without it.
Without skill50%
With skill59%
Checks it passed that the model alone missed
Recommends breaking down spend by region, top resources, and tags to isolate the driver
Recommends verifying that cost trends return toward baseline in the subsequent billing window
Recommends validating whether spend was planned or unexpected before altering data replication architectures
and 1 more in the full report
4 fixed9 passed either way7 still failing2 broke
Pass-rate lift
+9 pp
Tokens
-31%
Agent turns
0%
Response time
-47%
Model gemini-3.6-flashMeasured 8/3/2026v113 of 22 cases passed with skill
Tokens, turns and time describe resource use, not a dollar price.
See it in action
One of 4 test cases the skill fixed. Same model, same prompt, with and without the skill.
1 of 4
Prompt
An automated EBS snapshot backup script created thousands of uncompressed volume snapshots without cleanup, adding $5,000 daily to our AWS bill. What decision points apply?
What was checked, without and with the skill
Checked
Without skill
With skill
Recommends auditing snapshot retention schedules, object lifecycle policies, and log retention under storage decision points
Missed
Passed
case-19 · gemini-3.6-flash · 8/3/2026 · answers are capped and redacted
What's inside
The instructions your model reads when the skill loads.
SKILL.md42 linessingle file
When to use
Use when daily cost increases significantly versus baseline.
Preconditions
You can access billing/cost breakdown and recent change/deploy history.
Procedure
Confirm the spike magnitude and when it started.
Break down by service, region, and top resources/tags.
Identify the change driver: deploy, scaling, traffic, misconfiguration, abuse.
Validate whether spend is expected (planned test) or unexpected.
Choose the safest mitigation: pause non-critical workloads, cap scaling, fix loops.
Decision points
Compute spike: check autoscaling, instance sizes, and runaway jobs.