agentleFS
4444J994444J99/a-i--skills/skills/development/gcp-resource-optimizer/SKILL.md

gcp-resource-optimizer

Skill17 starsChanged 5 months agoApache-2.0Archived repositoryView on GitHub

Optimize Google Cloud Platform resource allocation and manage cloud credits efficiently. Use when planning GCP deployments, analyzing cloud spend, maximizing value from expiring credits, right-sizing instances, or designing cost-effective architectures. Triggers on GCP cost optimization, credit management, resource allocation planning, or cloud budget concerns.

1 other project uses a version of this.

The file

We split this skill into the rules it gives an agent

1 rule. About 0 are common, 1 is distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    When managing expiring credits:

    line 18

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

GCP Resource Optimizer

Maximize value from GCP resources and credits through strategic allocation.

Credit Burn Strategy

Credit Expiration Planning

When managing expiring credits:

  1. Audit current usage: gcloud billing accounts describe ACCOUNT_ID
  2. Calculate burn rate: Total credits ÷ Days remaining = Required daily spend
  3. Identify high-value uses: What creates lasting value vs. ephemeral compute?

High-Value Credit Uses

Lasting value (prioritize):

  • Training ML models (artifacts persist)
  • Building container images
  • Generating datasets
  • Running batch processing on accumulated work

Ephemeral (use strategically):

  • Compute instances (gone when shut down)
  • Development environments
  • Testing infrastructure

Cost Optimization Patterns

Compute Engine

Right-sizing instances:

# Check recommendations
gcloud recommender recommendations list \
  --project=PROJECT_ID \
  --location=ZONE \
  --recommender=google.compute.instance.MachineTypeRecommender

Cost-effective machine types:

NeedRecommendedWhy
General workloade2-mediumBest price/performance
Memory-intensiven2-highmemBetter RAM ratio
CPU burste2-micro/smallBurstable, cheap
ML trainingn1 + GPURequired for accelerators
Spot-tolerantSpot VMs60-91% discount

Preemptible/Spot VMs:

  • 60-91% cheaper than standard
  • Can be terminated with 30s notice
  • Good for: batch jobs, fault-tolerant workloads, development
  • Bad for: production, stateful services

Cloud Run

Optimizing Cloud Run:

# Minimize cold starts and costs
spec:
  template:
    spec:
      containerConcurrency: 80  # Maximize requests per instance
      timeoutSeconds: 300
    metadata:
      annotations:
        autoscaling.knative.dev/minScale: '0'  # Scale to zero
        autoscaling.knative.dev/maxScale: '10'  # Cap costs
        run.googleapis.com/cpu-throttling: 'true'  # CPU only when processing

Cloud Storage

Storage class optimization:

ClassUse CaseCost/GB/mo
StandardFrequent access~$0.020
NearlineMonthly access~$0.010
ColdlineQuarterly access~$0.004
ArchiveYearly access~$0.0012

Lifecycle rules:

{
  "lifecycle": {
    "rule": [
      {
        "action": {"type": "SetStorageClass", "storageClass": "NEARLINE"},
        "condition": {"age": 30}
      },
      {
        "action": {"type": "SetStorageClass", "storageClass": "COLDLINE"},
        "condition": {"age": 90}
      },
      {
        "action": {"type": "Delete"},
        "condition": {"age": 365}
      }
    ]
  }
}

BigQuery

Cost control:

-- Set maximum bytes billed
#standardSQL
-- @maximumBytesBilled 10000000000
SELECT * FROM dataset.table

Partitioning for cost reduction:

CREATE TABLE dataset.table
PARTITION BY DATE(timestamp_column)
CLUSTER BY user_id
AS SELECT * FROM source_table

Budget Alerts

Set up budget alerts:

gcloud billing budgets create \
  --billing-account=BILLING_ACCOUNT_ID \
  --display-name="Monthly Budget" \
  --budget-amount=100USD \
  --threshold-rule=percent=50 \
  --threshold-rule=percent=90 \
  --threshold-rule=percent=100

Resource Cleanup

Find Unused Resources

# Unused disks
gcloud compute disks list --filter="NOT users:*"

# Unused IPs
gcloud compute addresses list --filter="status=RESERVED"

# Idle VMs (by CPU)
gcloud monitoring time-series list \
  --filter='metric.type="compute.googleapis.com/instance/cpu/utilization"' \
  --interval="start=2024-01-01T00:00:00Z"

Cleanup Script

#!/bin/bash
# cleanup_unused.sh - Review before running!

# List (don't delete) unused resources
echo "=== Unused Disks ==="
gcloud compute disks list --filter="NOT users:*" --format="table(name,zone,sizeGb)"

echo "=== Reserved IPs ==="
gcloud compute addresses list --filter="status=RESERVED" --format="table(name,region,address)"

echo "=== Snapshots older than 30 days ==="
gcloud compute snapshots list --filter="creationTimestamp<$(date -d '30 days ago' -Iseconds)" --format="table(name,diskSizeGb,creationTimestamp)"

Architecture Patterns for Cost

Serverless-First

Request → Cloud Run → Firestore → Done
         (scales to zero)  (pay per op)

vs.

Request → GKE → Cloud SQL → Done
         (always running)  (always running)

Batch Processing

Pub/Sub → Cloud Functions → BigQuery (batch load)
                           (cheaper than streaming)

Development vs Production

Dev environment:

  • Spot/preemptible VMs
  • Smaller machine types
  • Scale-to-zero services
  • Shared resources

Prod environment:

  • Committed use discounts (1-3 year)
  • Right-sized dedicated instances
  • Redundancy only where needed

Monitoring Setup

# Enable billing export to BigQuery
gcloud beta billing accounts describe ACCOUNT_ID

# Query costs
#standardSQL
SELECT
  service.description,
  SUM(cost) as total_cost
FROM `project.dataset.gcp_billing_export_v1_*`
WHERE _PARTITIONTIME >= TIMESTAMP_SUB(CURRENT_TIMESTAMP(), INTERVAL 30 DAY)
GROUP BY 1
ORDER BY 2 DESC

References

  • references/pricing-cheatsheet.md - Quick pricing reference
  • references/cost-queries.md - BigQuery cost analysis queries

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.