nw-discuss
PO-led product conversation, visible feedback increments and human-readable briefs.
Install / Use
npx skills add nWave-ai/nWave --skill nw-discussInstalls into whichever agent you are using.
SKILL.md
Installable skill definition
Quality Score
Category
Development & EngineeringSupported Platforms
Our assessment of nw-discuss
nw-discuss scores 89/100 on our quality scale, 1725th of 4,702 Development & Engineering skills we index (top 37%).
Its SKILL.md is 29 KB long, well organised into 37 sections with 6 code examples: a thorough specification that gives an agent plenty to work with.
It has 617 GitHub stars, a meaningful sign that others use it.
Maintenance, license and trust
- The repository was last updated about 2 months ago, so nw-discuss is actively maintained.
- Our last check on 2026-10-31 found the source still online.
- It is released under the MIT license, a permissive license that allows use, modification and commercial use with attribution.
- Its trust signals score 100/100, with no cautions. These come from repository metadata, not a code audit — read the skill file before letting an agent act on it.
Safety scan
No issues foundOur scan of the whole file found no instruction hijacking, hidden characters, credential access, data exfiltration or destructive commands. An AI review of the same text found nothing harmful.
AI review by kimi-k2.7-code on 2026-10-08. Automated pattern scan on 2026-10-08. It catches known dangerous patterns, not every risk — read a skill before letting an agent act on it.
nw-discuss compared with similar skills
All 4 of these similar skills score higher than nw-discuss; compare them before choosing.
| Skill | Score | Stars | Updated | Format |
|---|---|---|---|---|
| nw-discuss (this skill)by nWave-ai | 89 | 617 | 2mo ago | SKILL.md |
| taste-skillby Leonxlnx | 100 | 117.6k | 9d ago | CLAUDE.md |
| ai-job-searchby ai-job-search-dev | 100 | 72.3k | today | CLAUDE.md |
| claude-howtoby luongnv89 | 100 | 47.4k | 3d ago | CLAUDE.md |
| lobehubby lobehub | 100 | 85.7k | today | MCP Server |
Frequently asked questions
- How do I install nw-discuss?
- Run
npx skills add nWave-ai/nWave --skill nw-discuss. The install tabs above show the steps for each supported agent. - Which AI agents does nw-discuss work with?
- It is written for Universal, as a SKILL.md file. Other agents that read the same format can often use it too.
- Is nw-discuss safe to use?
- Our scan of the whole file found no instruction hijacking, hidden characters, credential access, data exfiltration or destructive commands. An AI review of the same text found nothing harmful. It is MIT-licensed and scores 100/100 on trust signals. Skills are instructions an agent will follow, so read the file before installing it and do not approve commands you do not understand.
- Is nw-discuss still maintained?
- The repository was last updated about 2 months ago, so nw-discuss is actively maintained.
Skill content
View source on GitHubname: nw-discuss description: "Conducts Jobs-to-be-Done analysis, UX journey design, and requirements gathering through interactive discovery. Use when starting feature analysis, defining user stories, or creating acceptance criteria." user-invocable: true argument-hint: '[feature-name] - Optional: --phase=[jtbd|journey|requirements] --interactive=[high|moderate] --output-format=[md|yaml]'
NW-DISCUSS: Jobs-to-be-Done Analysis, UX Journey Design, and Requirements Gathering
Wave: DISCUSS (wave 2 of 6) | Agent: Luna (nw-product-owner) | Command: /nw-discuss
Overview
Execute DISCUSS wave through Luna's integrated workflow: JTBD analysis|UX journey discovery|emotional arc design|shared artifact tracking|requirements gathering|user story creation|acceptance criteria definition. Luna uncovers jobs users accomplish, maps to journeys and requirements, handles complete lifecycle from user motivations through DoR-validated stories ready for DESIGN. Establishes the acceptance-test foundation.
For greenfield projects (no src/ code, no docs/feature/ history), Luna proposes Walking Skeleton as Feature 0.
Output Tiers (per D2)
Provenance: feature lean-wave-documentation — D2 (schema-typed sections), D10 (one-line expansion descriptions), DDD-7 (DISCUSS pilot wave), D6 (install-time pedagogical prompt). Tier-1 [REF] sections (always emitted) + Tier-2 EXPANSION CATALOG items (lazy, on-demand) are the two output bands. Full contract: nWave/skills/nw-density-resolution-contract/SKILL.md.
Tier-1 [REF] — always emitted
Under ## Wave: DISCUSS / [REF] <Section> headings:
- Persona ID — one-line user identifier mapped to the journey
- JTBD one-liner — single-sentence Job-to-be-Done statement
- Locked decisions — D-numbered design decisions with verdicts
- User stories with elevator pitches — every story has Before/After/Decision-enabled triplet
- Acceptance criteria (ACs) — testable, embedded per story
- Definition of Done (DoD) — 9-item checklist
- Out-of-scope — explicit non-goals
- WS strategy — A/B/C/D per Mandate 5
- Driving ports — inbound surfaces (CLI, skill, HTTP)
- Pre-requisites — dependencies on prior waves or features
Tier-2 EXPANSION CATALOG — lazy, on-demand (per D10)
Rendered under ## Wave: DISCUSS / [WHY|HOW] <Section> only when requested via --expand <id> (DDD-2), the wave-end menu (expansion_prompt = "ask" or "ask-intelligent"), mode = "full" auto-expansion, or an ad-hoc user request mid-session.
| Expansion ID | Tier label | One-line description |
|---|---|---|
| jtbd-narrative | [WHY] | Full JTBD analysis: job dimensions (functional/emotional/social), four forces, opportunity scores |
| persona-narrative | [WHY] | Extended persona: goals, frustrations, mental model, vocabulary glossary |
| alternatives-considered | [WHY] | Decision rationale: alternatives weighed and rejected per locked decision |
| migration-playbook | [HOW] | Step-by-step migration guide for users on a prior version |
| journey-deep-dive | [HOW] | Full UX journey: emotional arc, shared artifacts registry, error-path map |
| gherkin-scenarios | [HOW] | Generated Gherkin scenarios covering happy path and key error paths |
| reviewer-findings-trace | [WHY] | R1-R10 reviewer findings chain with verdicts and how each landed in D1-D10 |
| expansion-catalog-rationale | [WHY] | Why this set of expansions, why these defaults, why D10 enforces one-line descriptions |
Density resolution (per D12)
Call resolve_density(global_config) from scripts/shared/density_config.py after reading ~/.nwave/global-config.json (missing/malformed = empty dict). Returns mode ("lean" | "full") + expansion_prompt ("ask" | "ask-intelligent" | "always-skip" | "always-expand" | "smart") per the D12 cascade (resolver-internal, DDD-5 — do NOT replicate locally). DISCUSS hard default is lean+ask-intelligent per Decision 4 (2026-04-28). Branch on density.mode (lean = Tier-1 only; full = Tier-1 + all Tier-2) and at wave end on density.expansion_prompt. Full cascade detail, branch semantics, ad-hoc override workflow: nWave/skills/nw-density-resolution-contract/SKILL.md.
Trigger detection (ask-intelligent mode, per Decision 4)
DISCUSS-specific extension on top of the shared contract. When expansion_prompt = "ask-intelligent", evaluate ALL triggers below against the wave artifacts produced so far. Each trigger that fires contributes its suggested expansion to a scoped menu. If NO trigger fires, emit no menu — strict lean output.
| Trigger | Detection criterion | Suggested expansion |
|---------|--------------------|--------------------|
| AC ambiguity | ≥2 user stories share an AC where reasonable readers could disagree on the outcome | gherkin-scenarios |
| Cross-context complexity | Feature touches ≥3 bounded contexts (per DDD glossary) OR ≥3 distinct technologies | alternatives-considered |
| Multi-stakeholder need | ≥3 distinct personas referenced across the user stories | persona-narrative |
| Compliance / regulatory | ACs reference regulatory terms (GDPR, HIPAA, SOX, audit, retention, encryption, PII, data residency) | migration-playbook (data migration) OR journey-deep-dive (user-facing) |
| WS strategy = D | Walking Skeleton strategy is "Configurable" (env-switching) | alternatives-considered |
Menu when 1+ trigger fires: Suggested expansions for this feature (triggered by: {trigger names}): - {id}: {description} ... Apply? [Y/n/all/none/custom]. Do NOT show the generic 8-item Tier-2 catalog in ask-intelligent mode — only triggered items. Ad-hoc override path ("expand <X>") still works for any catalog item. Telemetry: one event per scoped-menu choice; when NO trigger fires, one choice = "skip" event with expansion_id = "*" records the silent-lean opportunity.
Telemetry (per D4 + DDD-6)
Every expansion choice emits a DocumentationDensityEvent (dataclass at src/des/domain/telemetry/documentation_density_event.py) via event.to_audit_event() → JsonlAuditLogWriter().log_event(...). Schema fields per D4: feature_id, wave, expansion_id, choice, timestamp. For this wave the schema declares "wave": "DISCUSS". Use helper scripts/shared/telemetry.py:write_density_event(...) — do NOT write JSONL directly.
Wave-specific signal: feeds DDD-7 pilot success metric (4) — "downstream agent regression — DESIGN consumes lean DISCUSS feature-delta.md and produces no --expand invocation". ask-intelligent emission rules: one expand event per scoped-menu acceptance; one skip event for no-trigger silent-lean; one skip event for triggers fired but user declined. Full emission rules + per-mode patterns: nWave/skills/nw-density-resolution-contract/SKILL.md.
Interactive Decision Points
Decision 1: Feature Type
Question: What type of feature is this? Options:
- User-facing -- UI/UX functionality visible to end users
- Backend -- APIs, services, data processing
- Infrastructure -- DevOps, CI/CD, tooling
- Cross-cutting -- Spans multiple layers (auth, logging, etc.)
- Other -- user provides custom input
Decision 2: Walking Skeleton
Question: Should we start with a walking skeleton? Options:
- Yes -- recommended for greenfield projects
- Depends -- brownfield; Luna evaluates existing structure first
- No -- feature is isolated enough to skip
Decision 3: UX Research Depth
Question: Priority for UX research depth? Options:
- Lightweight -- quick journey map, focus on happy path
- Comprehensive -- full experience mapping with emotional arcs
- Deep-dive -- extensive user research, multiple personas, edge cases
Decision 4: JTBD Analysis
Question: Include Jobs-to-be-Done analysis? Options:
- Yes -- mandatory by default. Every user-facing story must trace to a
job_idindocs/product/jobs.yaml. Stories without job traceability fail Definition of Ready. - No (infrastructure-only escape valve) -- only permitted when the feature is a pure internal change (e.g. rename internal module, refactor build script) with no user-visible behavior. Requires
job_id: infrastructure-onlyAND ainfrastructure_rationalefield on every story explaining why no user job applies. Reviewer will reject this option for any feature that touches user-facing surfaces.
Default: 1 (Yes). Rationale: STANDING rule "Tech-surface vs value-outcome backlog anti-pattern" (2026-04-24) — epics with tech-surface children but no JTBD framing fail to converge on done-state. Default-on JTBD enforces value-outcome framing at PO level.
Prior Wave Consultation
Before beginning DISCUSS work, read SSOT and prior wave artifacts:
- SSOT (if
docs/product/exists):docs/product/journeys/{name}.yaml— existing journey to extend (if applicable)docs/product/jobs.yaml— validated jobs and opportunity scoresdocs/product/vision.md— product vision
- Project context:
docs/project-brief.md|docs/stakeholders.yaml - DISCOVER artifacts: Read
docs/feature/{feature-id}/discover/(if present) - DIVERGE artifacts: Read
docs/feature/{feature-id}/diverge/recommendation.mdandjob-analysis.md(if present — job is already validated, do not re-run JTBD)
Migration gate: If docs/product/ does not exist but docs/feature/ has existing features, STOP. The project has old-model features that should be migrated to SSOT before new waves run. Guide the user to docs/guides/migrating-to-ssot-model/README.md and complete the migration first. If docs/product/ does not exist and no old features exist (greenfield), DISCUSS will bootstrap it.
DISCUSS follows DISCOVER and optionally DIVERGE — reading SSOT first ensures continuity with prior features, then prior wave artifacts ground requirements in evidence.
READING ENFORCEMENT: You MUST read every file listed in Prior Wave Consultation above using the Read tool before proceeding. After reading, output a confirmation checklist (✓ {file} for each read, ⊘ {file} (not found) for missing). Do NOT skip files that exist — skipping causes requirements disconnected from evidence.
After reading, check whether any DISCUSS decisions would contradict DISCOVER evidence. Flag contradictions and resolve with user before proceeding. Example: DISCOVER found "users don't want automation" but DISCUSS story assumes "automated workflow" — this must be resolved.
Document Update (Back-Propagation)
When DISCUSS decisions change assumptions established in DISCOVER:
- Document change — Add a
## Changed Assumptionssection at the end of the affected DISCUSS artifact. Gate: section exists in artifact. - Reference original — Quote the original DISCOVER document and the original assumption verbatim. Gate: source document and quote both present.
- State new assumption — State the new assumption and rationale for the change. Gate: rationale is explicit.
- Preserve DISCOVER — Do NOT modify DISCOVER documents directly. Gate: DISCOVER documents unchanged.
Agent Invocation
@nw-product-owner
IF Decision 4 = Yes (default): Execute *jtbd-analysis for {feature-id}, then *journey informed by JTBD artifacts, then *story-map, then *gather-requirements with outcome KPIs. Every user story must include a job_id field traceable to docs/product/jobs.yaml.
IF Decision 4 = No (infrastructure-only escape valve): Execute *journey for {feature-id}, then *story-map, then *gather-requirements with outcome KPIs. Every story must use job_id: infrastructure-only AND include an infrastructure_rationale field. Reviewer rejects this branch for any user-facing feature.
Context files: see Prior Wave Consultation above + project context files.
Configuration:
- format: visual | yaml | gherkin | all (default: all)
- research_depth: {Decision 3} | interactive: high | output_format: markdown
- elicitation_depth: comprehensive | feature_type: {Decision 1}
- walking_skeleton: {Decision 2}
- outp
Truncated for display — read the full file on GitHub.
Related Skills
taste-skill
117.6kAgent skills that give coding agents taste: art direction, reference images and a browser check instead of generic AI pages
ai-job-search
72.3kThe job search that runs on your machine. AI job application framework built on Claude Code: evaluate postings, tailor CVs, write cover letters, prep interviews. Fork it and own it.
claude-howto
47.4kA visual, example-driven guide to Claude Code — from basic concepts to advanced agents, with copy-paste templates that bring immediate value.
lobehub
85.7k🤯 LobeHub - Super-Agents on Your Team. Add super-agents to the chats your team already uses.They take on work by themselves and get better every week.
Languages
Trust signals
From repository metadata: license, adoption, age and documentation. Not a code audit — see the Safety scan above for what the skill file itself contains.
