nw-ab-critique-dimensi…
Review dimensions for validating agent quality - template compliance, safety, testing, and priority validation
4-phase discovery workflow with decision gates, phase transitions, success metrics, and state tracking
$ npx -y skills add nWave-ai/nWave --skill nw-discovery-workflow --agent claude-codeHow it fires
How this skill gets triggered: by you, by Claude, or both.
/nw-discovery-workflowContext preview
The summary Claude sees to decide when to auto-load this skill.
4-phase discovery workflow with decision gates, phase transitions, success metrics, and state tracking
name: nw-discovery-workflow description: 4-phase discovery workflow with decision gates, phase transitions, success metrics, and state tracking user-invocable: false disable-model-invocation: true
PHASE 1 PHASE 2 PHASE 3 PHASE 4
Problem Validation Opportunity Mapping Solution Testing Market Viability
| | | |
v v v v
"Is this real?" "Which matters?" "Does it work?" "Viable business?"Duration: 1-2 weeks | Min interviews: 5 | Techniques: Mom Test, Job Mapping Core question: Is this a real problem worth solving?
Duration: 1-2 weeks | Min interviews: 10 cumulative | Techniques: OST, Opportunity Algorithm Core question: Which problems matter most?
Duration: 2-4 weeks | Min interviews: 5 per iteration | Techniques: hypothesis testing, prototypes Core question: Does our solution actually work?
Duration: 2-4 weeks | Min interviews: 5 + stakeholders | Techniques: Lean Canvas, 4 Big Risks Core question: Can we build a viable business?
Proceed: 5+ confirm pain + willingness to pay | Pivot: problem differs from expected | Kill: <20% confirm
Proceed: top 2-3 score >8 (max 20) | Pivot: new opportunities discovered | Kill: all low-value Scoring: Score = Importance + Max(0, Importance - Satisfaction). Each 1-10. >8 = high importance with satisfaction gap.
Proceed: >80% task completion, usability validated | Pivot: needs refinement | Kill: fundamental blocks
Proceed: all 4 risks addressed, model validated | Pivot: model needs adjustment | Kill: no viable model
| Metric | Target | |--------|--------| | Problem confirmation | >60% (3+ of 5) | | Frequency | Weekly+ | | Current spending | >$0 on workarounds | | Emotional intensity | Frustration evident |
Done when: 5+ interviews, >60% confirmation, articulated in customer words, 3+ examples. Threshold: 60% aligns with Mom Test -- 3/5 consistent signals = proceed, <20% = kill. Combined with qualitative markers.
| Metric | Target | |--------|--------| | Opportunities identified | 5+ distinct | | Top scores | >8 / max 20 | | Job step coverage | 80%+ | | Strategic alignment | Stakeholder confirmed |
Done when: OST complete, top 2-3 prioritized, team aligned.
| Metric | Target | |--------|--------| | Task completion | >80% | | Value perception | >70% "would use/buy" | | Comprehension | <10 sec to understand | | Key assumptions validated | >80% proven |
Done when: 5+ users per iteration, core flow usable, value + feasibility confirmed.
| Metric | Target | |--------|--------| | Four big risks | All green/yellow | | Channel validated | 1+ viable | | Unit economics | LTV > 3x CAC (estimated) | | Stakeholder signoff | Legal, finance, ops |
Done when: Lean Canvas complete, all risks acceptable, go/no-go documented.
current_phase: "1|2|3|4" phase_started: "ISO timestamp" interviews_completed: "count by phase" assumptions_tracked: "list with risk scores" opportunities_identified: "list with scores" decision_gates_evaluated: "G1|G2|G3|G4 status" artifacts_created: "list of file paths"
| Transition | Gate | Requirements | |-----------|------|-------------| | 1 to 2 | G1 | 5+ interviews, >60% confirmation, customer words, 3+ examples | | 2 to 3 | G2 | OST complete, top 2-3 identified, scores >8, team alignment | | 3 to 4 | G3 | 5+ users tested, >80% task completion, core flow usable, validated | | 4 to handoff | G4 | Lean Canvas complete, 4 risks acceptable, go/no-go, stakeholder sign-off |
AI agents that guide you from idea to working code, with human judgment at every gate. nWave runs inside Claude Code. It breaks feature delivery into seven waves (discover, diverge, discuss, design, devops, distill, deliver).
Repo: nWave-ai/nWave
Review dimensions for validating agent quality - template compliance, safety, testing, and priority validation
Review dimensions for validating agent quality - template compliance, safety, testing, and priority validation
Review dimensions for acceptance test quality - happy path bias, GWT compliance, business language purity, coverage completeness, walking skeleton…
Detailed 5-phase workflow for creating agents - from requirements analysis through validation and iterative refinement
5-layer testing approach for agent validation including adversarial testing, security validation, and prompt injection resistance
Architectural style selection decision matrices, trade-off analysis, structural enforcement rules, and combination patterns. Load when choosing or evaluating…