nWave-ai/nWave / nWave/skills/nw-tdd-methodology
nw-tdd-methodology
VerifiedDeep knowledge for Outside-In TDD - double-loop architecture, ATDD integration, port-to-port testing, walking skeletons, and test doubles policy
Outside-In TDD Methodology
Lean composing core. The deep knowledge is decomposed into seven one-job-one-trigger modules; five ship as skills under the nw-tdd-methodology-* prefix and the other two are summarised in this core. This core keeps the name (shared — the crafters, acceptance-designer, distill, and the test-optimizer all reference it) and routes to the module whose trigger fires. Load the module(s) for the job in play; do not re-inline their content here.
Composition — module → trigger
| Module | Job | Trigger (when to load) |
|---|---|---|
| (summarised below) | the full TDD cycle | the invocation owns test authoring and runs RED→GREEN→COMMIT; production-only craft against an immutable oracle does not trigger this job |
| (summarised below) | workflow shape | deciding the development-workflow shape — double-loop, Outside-In vs Inside-Out, lightweight ATDD, BDD, Bache workflow, unit-of-behavior framing |
nw-tdd-methodology-port-to-port | test boundary | deciding what a test enters/asserts on — port-to-port discipline, the layer-specific Universe, refactoring-resilience, hexagonal per-layer strategy |
nw-tdd-methodology-test-doubles | choose a double | choosing/building a test double — Meszaros taxonomy, classical-vs-mockist, mock-only-at-port-boundaries policy, doubles-must-validate-inputs contract |
nw-tdd-methodology-walking-skeleton | build a WS | building/validating a walking skeleton — WS protocol, feature-level skeleton with incremental slice coverage, Mandate 5 adapter-strategy decision tree, Mandate 6 adapter real-I/O, adapter-integration RED-phase semantics |
nw-tdd-methodology-paradigm | write a unit/acceptance test | writing a unit or acceptance test — property-based + state-delta mandate, applicability matrix, debt-payoff curve, delta-first trigger/bypass |
nw-tdd-methodology-pbt-deep | write a PBT (deep) | writing a property-based test needing deep mechanics — A13/P6 stateful preconditions, four property-finding strategies, two shrinking mechanisms, targeted/search-based PBT |
How to use
- Identify the job in play (which trigger fires).
- Read
~/.claude/skills/nw-tdd-methodology-{module}/SKILL.mdfor that job unless its body is already preloaded or the job is summarised below. - For an invocation owning the full test-authoring cycle, apply the cycle summary +
-paradigm; add other modules when their triggers apply. Managed production-only crafters implement against the immutable ATD-owned oracle and report oracle defects to its owner; the cycle summary grants no test-authoring or commit authority.
The seven modules together cover the full methodology.
Jobs summarised in this core
- Cycle — 3-phase RED/GREEN/COMMIT canon. A new test must fail for the right reason before any production code is written; GREEN matches the public contract; no production code without a requiring test; after GREEN verify the new code is wired into the real driving port. Code examples follow the project's own language.
- Workflow shape — double-loop TDD: an outer acceptance test through the driving port stays red while inner unit-level cycles drive the implementation green. Choose Outside-In (acceptance first, collaborators discovered) or Inside-Out (domain first) by how much of the design is known; keep the unit of behavior an observable outcome, not a function or class.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| finishing-a-development-branchobra/superpowersUse when implementation is complete, all tests pass, and you need to decide how to integrate the work | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| systematic-debuggingobra/superpowersUse when encountering any bug, test failure, or unexpected behavior, before proposing fixes | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| test-driven-developmentobra/superpowersUse when implementing any feature or bugfix, before writing implementation code | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| skill-creatoranthropics/skillsCreate new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, edit, or optimize an existing skill, run evals to test a skill, benchmark skill performance with variance analysis, or optimize a skill's description for better triggering accuracy. | Stars: 197k | Updated | Claude CodeCodex | Verified | Install |