Finops cost optimization
Skill SWEStash/swe-workflow-skills/plugins/strategy/skills/finops-cost-optimization
Understand and reduce cloud spend — cost visibility and allocation (tagging, ownership), unit economics, rightsizing, autoscaling, commitment discounts (reserved instances, savings plans), storage/egress traps, architecture cost review, budgets and anomaly alerts. Triggers: cloud costs, AWS bill, cut costs, cost optimization, FinOps, why is our bill so high, rightsizing, reserved instances, savings plan, egress fees, cost per customer. Whether to build or buy a capability → build-vs-buy; making code faster → performance-optimization.From its SKILL.md
npx -y skills add SWEStash/swe-workflow-skills --skill finops-cost-optimizationAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
One thing to look at
- 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
SKILL.md
5.6 KB, ~1.2k tokens by cl100k_base, as published. Nobody here has run it
FinOps & Cost Optimization
Treat cloud cost as an engineering metric — observable, attributed, and
optimized with the same rigor as latency. The failure mode this skill counters:
nobody owns the bill, so it only gets attention as a quarterly surprise, and
the response is a panicked freeze instead of systematic reduction. Boundaries:
performance-optimization makes code faster (sometimes cheaper as a side
effect); build-vs-buy decides whether to run the capability at all; this
skill optimizes what you do run.
Workflow
Step 1: Visibility Before Optimization
You can't cut what you can't attribute. Establish first:
- Allocation: tag/label resources by service, team, and environment; measure the untagged remainder and drive it down — an 40%-unattributed bill makes every later step guesswork.
- The big rocks: rank spend by service and by resource type. Cloud bills follow a power law; the top 5 line items usually carry most of the total. Optimizing anything outside them is procrastination with charts.
- Trend vs events: is spend growing with usage (fine, if unit costs hold), or did it step-change with a deploy (an incident with a dollar signature)?
Pull real billing data (cost explorer exports, infracost on IaC diffs, the
provider CLIs) — never optimize from memory of last quarter's bill.
Step 2: Define Unit Economics
Total spend is noise; cost per unit of value — per customer, per request,
per GB processed, per model inference — is signal. A bill that doubles while
customers triple is a win. Pick 1–3 unit metrics with the team, compute them
from the Step-1 allocation, and make them the number optimization is judged by.
This is also what connects engineering to the business conversation
(metrics-and-okrs).
Step 3: Harvest the Waste (no architecture changes)
The reliable first wins, in typical order of return:
- Orphans: unattached volumes/IPs, forgotten environments, snapshots-forever, idle dev clusters running nights and weekends (schedule them off).
- Rightsizing: instances/databases provisioned for imagined peak — check actual utilization (single-digit CPU on a large instance is a purchase order for the cloud provider, not capacity planning).
- Storage classes & lifecycle: hot-tier data nobody has read in a year; logs and backups with no expiry — lifecycle policies are one-line fixes.
- Non-prod parity: staging doesn't need production's instance sizes, replica counts, or retention.
Step 4: Buy Commitment for the Stable Baseline
After rightsizing (never before — you'd lock in the waste): cover the stable, predictable baseline with reserved instances / savings plans / committed-use discounts, and keep the spiky remainder on-demand or spot (spot for interruption-tolerant work: batch, CI, stateless workers). Start conservative (cover well under 100% of the baseline) and ratchet up as forecasts prove out — over-commitment converts a variable cost into a fixed regret.
Step 5: Review the Architecture's Cost Shape
Some spend is structural, and only architecture changes move it:
- Egress and cross-zone/region chatter — the classic invisible line item; co-locate chatty services, cache at the edge, compress.
- Per-invocation pricing at scale — serverless that grew past the crossover where instances are cheaper (and vice versa: idle instances that should be serverless).
- Managed-premium vs ops-burden — the managed service's markup is often
worth it (that's a
build-vs-buycall when it gets big). - Data pipelines — scan-based pricing (warehouse queries over unpartitioned
tables) and always-on streaming for hourly-batch needs (
data-pipeline-design).
Cost these with current provider pricing (verify live — pricing changes), and
record significant trade-offs as ADRs (architecture-design).
Step 6: Keep It Continuous
- Budgets + anomaly alerts per team/service, routed to the owning team — a cost regression is a defect, detected in days not quarters.
- Cost in the deploy loop: IaC cost diffs (e.g. infracost) in PRs makes cost visible at decision time.
- A monthly 30-minute review of the unit metrics and top movers beats an annual cost-cutting project.
Principles Applied
- Attribution before optimization — unowned costs never stay cut.
- Unit economics over totals — growth should raise the bill; only rising unit cost is a problem.
- Biggest line item first — a 10% cut of the top item beats a 90% cut of the tenth.
Cross-Skill References
build-vs-buy— whether to run the capability at all; managed-vs-self-hosted TCOperformance-optimization— efficiency work that reduces compute per requestarchitecture-design— recording cost-driven structural decisions as ADRsobservability-design— the metrics/alerting machinery cost signals ride ondata-pipeline-design— cost-shaping batch/streaming and warehouse workloadsmetrics-and-okrs— tying unit economics to business metrics
What ships with it: 1 file
4.0 KB alongside SKILL.md
evals/
- evals.json4.0 KB
Gives 0 of the 12 instructions most token cost skills give in ~1.2k tokens
Counted across 84 of the 95 authors here whose files we hold, read 2026-09-06
- Set explicit budget limits before processing batchesin 14 of 84, across 9 files
- Define model names as constants or configin 14 of 84, across 9 files
- Route simple tasks to cheaper modelsin 14 of 84, across 9 files
- Log model selection decisionsin 14 of 84, across 9 files
- Retry only transient errorsin 14 of 84, across 9 files
- Track spend with immutable cost recordsin 13 of 84, across 8 files
- Cache system prompts over 1024 tokensin 12 of 84, across 7 files
- Check the budget before each API callin 12 of 84, across 7 files
- Start with the cheapest modelin 9 of 84, across 8 files
- Back off exponentially between retriesin 9 of 84, across 8 files
- Review costs weeklyin 8 of 84, across 6 files
- Use reserved capacity for steady workloadsin 8 of 84, across 6 files
Said here and by no other author read
- Pull real billing data before optimizing
- Rank spend by service and resource type
- Pick unit metrics with the team
- Schedule idle dev clusters off
- Apply lifecycle policies to stale data
- Buy commitments only after rightsizing
Grouped from the skills themselves: near-identical wordings counted once, and counted by distinct author, so one author publishing three of these counts once. Length counted with cl100k_base; the agent that loads this file may tokenize it differently.