Owned local sample / Recorded 8 September 2026
See the bug.
Then inspect the evidence.
An invoice and shipping fixture with deliberately seeded faults. Compare the outcomes to see why one passing test cannot tell the whole story.
A passing test misses the shipping fault.
The fast-check property test passes, while unit, API, combinatorial property, browser and Gherkin checks fail. The sample deliberately changes the shipping boundary.
| METHOD | TASK RESULT | CHECKS PASSED | CHECKS FAILED |
|---|---|---|---|
| Unit tests | Failed | 3 | 1 |
| API checks | Failed | 2 | 1 |
| Property checks · combinations | Failed | 0 | 1 |
| Property checks · fast-check | Passed | 1 | 0 |
| Browser checks | Failed | 0 | 1 |
| Gherkin scenarios | Failed | 1 | 1 |
Six of twelve mutations survived in the fixed sample. This educational fixture was executed locally. It contains no customer data, cloud execution or model-generated results. These outcomes demonstrate the included tests and known faults; they do not establish customer results or live provider qualification.
Reproduce the unit examples yourself.
The evidence pack includes the specification, source, tests, measured outcomes, selected native reports and SHA256 checksums. Extract it and use Node 20 or later to run the included unit examples:
node --test sample/tests/rules.test.cjsThe exported files preserve the original internal product label, Tenhaw Quality. The name of the marketed product is still being decided.
Meet your AI testing workhorse
Bring your toughest code.
Set your highest bar.
See how the AI testing team would challenge your next release. We’ll scope the methods, automation and evidence around your software.
Discuss an evaluation. No purchase commitment.