Kubernetes Waste Prevention Platform

Find existing waste in your clusters. Block new waste in Pull Requests.

Most teams waste $30-80K/year on over-provisioned resources.
Wozz catches it before and after it hits production.

🔍

Audit (CLI)

Find existing waste

NEW
🛡️

Prevent (PR Bot)

Block new waste

Install GitHub Action
Open Source • Zero Setup • Privacy First • 2 Minutes

Works with

Kubernetes
AWS
GCP
Azure
kubectl

See What You're Wasting in Real-Time

This is what the terminal output looks like:

terminal — wozz audit
$ curl -sL wozz.io/audit.sh | bash
→ Connecting to cluster... ✓
→ Collecting resource data... ✓
→ Analyzing usage patterns... ✓
→ Calculating costs... ✓
WOZZ AUDIT RESULTS
💰 $52,000/year
Annual savings potential
1. Over-Provisioned Memory [HIGH]
• 12 pods affected
• Average: 6.5x over limit
• Savings: $2,800/month
2. Orphaned Load Balancers [MEDIUM]
• 2 idle LBs (90+ days)
• Savings: $1,200/month
📄 Full report: wozz-report.html
🌐 Visualize: https://wozz.io/analyze
NEW: GITHUB PR BOT

Stop Waste Before It Merges

Wozz automatically comments on Pull Requests when resource changes will increase cloud costs.

feat: scale up api-server memory
#247 • Open
🛡️
wozz-bot BOT commented 2 minutes ago
Wozz Cost Alert

This PR increases cloud costs by $576/year.

Resource Old New Cost Impact
api-server 4Gi 8Gi +$576/yr

💡 Recommendation: Review if this increase is necessary. Current usage shows 1.2Gi average.

👉 Analyze Risk with AI → • Docs
Add to Your Repo (2 min setup)

Runs on your GitHub Actions • Zero external access required

The #1 Source of K8s Waste

It's not poor planning. It's fear-based provisioning.

What Teams Set

resources:
  limits:
    memory: "8Gi"  # "Safe" default
    cpu: "2"
  requests:
    memory: "2Gi"
    cpu: "500m"
Cost: ~$70/month per pod
Why: Fear of OOM kills, CPU throttling, production incidents

What Actually Runs

$ kubectl top pod api-server

NAME         CPU    MEMORY
api-server   180m   450MB
Actual need: ~$8/month per pod
Waste: $62/month × 50 pods = $37K/year
20 out of 20 clusters
had this exact pattern: limits 5-10x higher than usage.

Defense Grid Architecture

Two layers of protection. Complete waste prevention.

🔍

Layer 1: Audit

Find existing waste

1

Scan Cluster

Reads pod configs and usage metrics via kubectl. Read-only access.

2

Calculate Waste

Compares limits vs actual usage. Identifies over-provisioned resources.

3

Show Savings

Instant terminal output + HTML report. Share with your team.

Use case: Weekly audits, cost reviews, compliance reporting

🛡️

Layer 2: PR Bot

Block new waste

1

Monitor PRs

Watches for changes to YAML manifests. Triggers on every pull request.

2

Analyze Diff

Detects resource changes. Calculates annual cost impact automatically.

3

Alert Team

Comments on PR if cost threshold exceeded. No surprises in production.

Use case: CI/CD pipeline, code review, cost governance

The Full Defense Grid

Audit finds what's already broken. PR Bot prevents new waste from merging. Together, they keep your clusters lean forever.

Zero external access Runs on your infrastructure Open source

What Teams Are Finding

$52K
Average savings found
38%
Average cluster waste
2 min
Time to first insight
✨

Ready to See
What's Really There?

Let's pull back the curtain together. See exactly where your infrastructure spend is going and discover the savings waiting to be found.

$ curl -sL wozz.io/audit.sh | bash
Get Started →

Open source • Runs locally • Zero data upload

Three Ways to Use Your Results

💻

Terminal Output

See savings immediately in your terminal. No waiting.

📄

HTML Report

Detailed report with charts opens automatically.

📊

Web Analyzer

Upload JSON for interactive charts. Try it →

Want AI-powered analysis & historical tracking?

Connect your repo to unlock smart ignore rules, trend analysis, and team collaboration features.

Connect GitHub Repo

No credit card required • 14-day free trial