nWave-ai/nWave / nWave/skills/nw-spike
nw-spike
VerifiedA bounded probe of one product, visual-experience or technical uncertainty; returns evidence to its owner without starting a wave or promoting code automatically.
NW-SPIKE
The LLM selects a probe when an unanswered question changes a product or design decision. This is a technique inside DISCUSS/DESIGN, not an extra mandatory wave. Honor an existing delegation or stop boundary; do not repeat permission requests.
- State one question, the evidence that would change the decision, and a bounded scope/time budget. Skip a probe that cannot inform a decision.
- Reuse relevant artifacts. For visual uncertainty, the PO supplies actor,
journey and feedback question;
nw-ux-designersupplies UX/UI competence. Choose sketch, wireframe or navigable prototype according to uncertainty. For technical uncertainty, select the applicable technical specialist. - Keep artifacts and findings in the caller-owned persistent workspace, outside
production paths; never rely on
/tmpfor valuable work. Use the project's native toolchain. Clearly label simulated data, mocked effects and absent integration. Prototype only enough behavior to answer the question. - Observe the result. For a visual prototype, show it to the human and retain actual feedback separately from model inference. Pending feedback stays open; an attractive page is not usability evidence. Do not force a binary verdict for a human preference or an unresolved question.
- Return evidence, limitations and the decision still needed to the PO or architect, who owns any resulting authority revision. Do not write wave authority or handover documents yourself.
No automatic promotion, commit or cleanup. A separately selected implementation may reuse prototype code after normal design, oracle and review work. A mocked prototype is not the feature's walking skeleton. Preserve useful evidence before any explicitly selected disposal; meaningful prototype changes require renewed feedback, not reuse of an old approval.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| receiving-code-reviewobra/superpowersUse when receiving code review feedback, before implementing suggestions, especially if feedback seems unclear or technically questionable - requires technical rigor and verification, not performative agreement or blind implementation | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| requesting-code-reviewobra/superpowersUse when completing tasks, implementing major features, or before merging to verify work meets requirements | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| using-git-worktreesobra/superpowersUse when starting feature work that needs isolation from current workspace or before executing implementation plans - ensures an isolated workspace exists via native tools or git worktree fallback | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| verification-before-completionobra/superpowersUse when about to claim work is complete, fixed, or passing, before committing or creating PRs - requires running verification commands and confirming output before making any success claims; evidence before assertions always | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |