Quality & Health

sate golden

Shipped

Identifies high-value tests — the ones that protect your business core (payments, billing, auth, subscriptions).

What it does

Scores every test on 5 factors (explicit @golden marker / @group golden, critical business patterns, complexity, historical stability, code importance) and surfaces the ones above a confidence threshold. Only scans actual test files (filename or content marker — never services, models, or routes).

Who runs this

→Engineering when picking which tests to prioritise during a refactor
→Incident reviewers asking "did we have a test that should have caught this?"

How to invoke

sate golden --project-root /path/to/laravel-app

Flags

--project-root <path>

Required.

--min-confidence <0..1>default: 0.4

Lower bound for inclusion.

--limit <N>default: 25

Cap display output.

--json

Machine-readable output.

What it produces

·spec/golden_tests.json — registry of golden tests with confidence scores and reasons

What it does NOT do

✖Does NOT change protection status in CI — just identifies candidates
✖Does NOT scan non-test files (services, models, routes)

Notes

On Maktabi flags 58 of 5,091 test files as golden — typically billing, subscription, payment, and webhook flows.

Status

Engine tests

111

Status

Shipped

Related pipelines

sate quality

Per-test quality scoring across 6 metrics: completeness, assertion strength, mutation resistance, structure, maintainability, reliability.

sate audit-all

The front door: runs every read-only pipeline against your project and writes both a developer report set and a non-technical owner summary.