Test integrity

Self-building AI code generation engine that generates application flows instead of implementing them. AGPL-3.0.From the repository description

Install

npx -y skills add lubochka/xiigen-mvp-engine --skill test-integrity

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

2 things to look at

20 days oldThe repository was created 20 days ago. New is not bad, but a brand new repository carrying a familiar-sounding name is the shape a typosquat arrives in, and there has been no time for anyone else to find a problem with it.
0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

1.7 KB, 282 tokens by cl100k_base, as published. Nobody here has run it

Use this when adding, changing, reviewing, or relying on tests.

Invoke for bug fixes, behavior changes, regression coverage, test failures, or any claim that tests prove completion.

Understand what behavior the test is supposed to protect before editing it.
Prefer testing public behavior and contracts over internal mechanics.
Do not weaken assertions to match broken behavior.
For bug fixes, reproduce the failure or document why reproduction is not available.
Run the smallest relevant test set after test edits and read the result.

Tests for services should verify DataProcessResult status and data/error semantics, not just non-null results.
Where relevant, test local async controls such as AbortSignal pass-through, NestJS provider or injection-token wiring, config helper defaults, logging-adjacent behavior without depending on log text, dynamic Record<string, unknown>/unknown payload handling, and buildSearchFilter/buildSearchFilterFlat empty-value behavior.

Do not delete, skip, or relax tests without explicit authority.
Do not claim full coverage from a narrow test.
Do not import source architecture, algorithms, design patterns, training, DPO, source domains, source classes, or source-specific paths.

Tests remain meaningful, behavior-aligned, and their results support the claims made.

Read from the repository

Just SKILL.md. No reference files, no scripts.