Automation Script Test Harness · Sample report

A real run: 11 cases, MAS 9.1 and 9.2, one script family.

The kit's own generated execution summary. Not a mockup — the same numbers you get running mvn verify yourself today. A clean result still carries two NOT ASSESSED findings. A lens the dictionary has no evidence for is stated. Never silently skipped past.

What the run found

22 executions, 11 cases, 2 lenses. 20 passed outright. 2 stated as not evidenced for their lens. Zero silent passes, zero invented failures. Source: tools/maximo-mockkit/reports/mockkit-execution-summary.md, generated 2026-09-06.

The two findings, as they appear

newFieldDefaultsShift — NOT ASSESSED · B2 dictionary observed only · MAS 9.1. ZZ_SHIFT is not in the MAS 9.1 dictionary. Skipped, not passed. The clean result on 9.2 is not read as covering 9.1 too.

shiftAttributeDoesNotExistOnNineOne — NOT ASSESSED · B2 dictionary observed only · MAS 9.2. The negative-case twin. ZZ_SHIFT exists on 9.2, so the "does not exist" assertion has nothing to test there.

What the other nine cases looked like

Set iteration. Domain validation. Status transitions. service.error() propagation from both the Jython and JavaScript launch points. The static MXServer entry point. All passed on both lenses. Passing on both is the unremarkable outcome a working script should produce. The two NOT ASSESSED rows above are what make that pass worth trusting — the report says where it did not look, instead of staying silent.

Downloadable today, and what a full bundle will add

The execution summary (Markdown) and the kit itself (source zip) are downloadable today, so you can reproduce this exact run. Brief/report/evidence pages and a lock file, in the shape described on the documentation page, are roadmap.