Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →nwave-ai/nw-tdd-review-enforcement
Test design mandate enforcement, test budget validation, TDD phase validation (3-phase canon per ADR-025), and external validity checks for the software crafter reviewer
Passed 21 of 22 test cases with the skill, and 10 of 22 without it.
and 12 more in the full report
11 fixed10 passed either way1 still failing0 broke
Tokens, turns and time describe resource use, not a dollar price.
One of 11 test cases the skill fixed. Same model, same prompt, with and without the skill.
A pull request has completed 3 review iterations with persistent gate failures on adapter integration tests. What is the required outcome decision?
| Checked | Without skill | With skill |
|---|---|---|
| Escalates the submission to tech lead after exceeding 2 review iterations. | Missed | Passed |
The instructions your model reads when the skill loads.
Domain knowledge for reviewing TDD implementations against 5 test design mandates, test budget, TDD phase compliance (3-phase canon per ADR-025), and external validity.
All assertions validate observable outcomes, never internal structure.
Observable: return values from driving ports | state changes via queries | side effects at driven port boundaries | exceptions from public API | business invariants
Violations: asserting private fields | verifying internal method call order | inspecting intermediate calculations | checking internal class instantiation
Severity: Blocker. Rewrite to assert observable outcomes only.
Zero unit tests of domain entities/value objects/services directly. Test indirectly through application service (driving port) tests.
Violations: imports domain entity (Order, Customer) | instantiates value object (Money, Email) | invokes domain service method
Exception: complex standalone algorithm with stable public interface (rare). Severity: Blocker. Delete domain tests, add application service test.
All unit tests enter through driving ports (application services, controllers, CLI handlers, event handlers). Never internal classes.
Detection: grep for internal imports (from domain.entity, from internal.validator).
Severity: Blocker. Rewrite through driving port.
Adapters have integration tests with real infrastructure (testcontainers, test servers). No mocked unit tests.
Violations: mocking IDbConnection | mocking SMTP client | stubs instead of real infrastructure Acceptable: in-memory implementations if behavior-complete. Severity: Blocker. Convert to integration test.
Input variations of same behavior use parametrized tests, not duplicates.
test cases passed, vs 10 / 22 without
More skills from the same source and nearby in the registry, with their headline benchmark lift.