wize-dev-kit 0.1.4 → 0.2.0

package/src/method-skills/2-plan-workflows/wize-ux-scenarios/workflow.md

@@ -4,45 +4,121 @@ name: UX Scenarios
 phase: 2-plan
 owner: wize-agent-ux-designer   # Mantis (absorbs WDS Freya)
 absorbs: "WDS Freya — Phase 3 (8-question dialog)"
-status: stub
+status: ready
 ---
 
 # UX Scenarios
 
-**Goal.** Produce the scenario map that anchors all UX work. Eight focused questions; one answer each.
+**Goal.** Produce a scenario map that anchors all UX work. Eight focused questions; one short answer each. Scenarios are pieces of behavior, not personas — they're how *this* user, in *this* state, tries to accomplish *that* JTBD.
+
+Mantis drives. Output lands in `.wize/planning/ux/ux-scenarios.md`. It's the bridge from PM intent (PRD) to designer concreteness (next workflow: `wize-ux-design`).
 
 ## Inputs
-- `.wize/planning/prd.md`
-- `.wize/planning/ux/trigger-map.md`
+
+- `.wize/planning/prd.md` (validated)
+- `.wize/planning/ux/trigger-map.md` (Pepper)
 - `.wize/planning/brief.md`
 
 ## Outputs
+
 - `.wize/planning/ux/ux-scenarios.md`
 
-## The 8 questions (WDS dialog)
+## Steps
+
+### 1. Identify the scenarios
+
+A scenario = one (user, state, JTBD) tuple where the product earns or loses. Aim for **3–8 scenarios** that, taken together, cover all backbone stories in the PRD.
+
+Rules:
+- One sentence each at this stage.
+- Verb-led. *"New manager onboards their first team."* Not *"Manager onboarding."*
+- Distinct. If two scenarios share trigger + state, collapse them.
+
+### 2. Run the 8-question dialog per scenario (WDS Freya)
+
+For each scenario, answer the eight questions. One paragraph per answer. No filler. Cite trigger-map row when relevant.
 
 1. **Who is the user, in one sentence?**
-2. **What state are they in when they arrive?** (emotional, contextual)
-3. **What are they trying to accomplish?** (job-to-be-done)
+   Role + context. *"A 38-year-old engineering manager at a 25-person SaaS, primarily on desktop, sometimes phone."*
+
+2. **What state are they in when they arrive?**
+   Emotion + situational context. *"It's Monday morning. They just got CTO's request to onboard 3 hires this sprint. They're alt-tabbing between Slack and Linear."*
+
+3. **What are they trying to accomplish?** (JTBD)
+   *"Add three teammates and assign each to the right project before the team standup at 10am."* Concrete.
+
 4. **What would they do without our product today?**
+   *"Slack screenshots of the project list to each new hire; have them sign up individually; cross-check who landed where."*
+
 5. **What's the moment of truth — when they decide it works?**
+   *"The first time they see all three names in the team list with the right project tags, in under 5 minutes."*
+
 6. **What's the failure mode — when do they walk away?**
+   *"If they can't tell whether the invite arrived. If the invite email looks like spam. If they have to leave the page to find someone's email."*
+
 7. **What does success look like to them, in their words?**
+   *"It just worked. I sent the invites, they signed up, they were on the right project."* (Imagine the future quote.)
+
 8. **What's the next thing we want them to do?**
+   *"Open the project where the first teammate landed and verify the access works."*
 
-## Output structure
+### 3. Cross-link to the PRD
+
+For each scenario, list the **backbone story IDs** + **AC IDs** it touches. Mantis won't write specs for a scenario the PRD doesn't cover (it'd be scope creep); Hill won't sign off on a story the scenarios don't surface (it'd be over-spec).
+
+### 4. Hand off to UX Design
+
+Mark `status: ready-for-design`. Wizer pings Mantis to start `wize-ux-design`, where each scenario becomes one or more page specs.
+
+## Output template
 
 ```markdown
-# UX Scenarios
+---
+status: ready-for-design
+owner: Mantis
+created: YYYY-MM-DD
+covers_prd_stories: [E01, E02, E04]
+---
+
+# UX Scenarios — {{project_name}}
 
-## Scenario 1: {{name}}
+## Scenario 1: New manager onboards their first team
 
-- Q1 …
-- Q2 …
-- …
+- **Trigger-map row:** 1 (Sign up), 2 (Invite first teammate)
+- **PRD stories:** E01, E02
+- **AC IDs touched:** AC-01-1, AC-01-2, AC-02-1, AC-02-3
 
-## Scenario N: …
+**Q1 — Who:** A 38-year-old engineering manager at a 25-person SaaS, primarily on desktop, sometimes phone.
+**Q2 — State:** It's Monday morning. They just got CTO's request to onboard 3 hires this sprint. They're alt-tabbing between Slack and Linear.
+**Q3 — JTBD:** Add three teammates and assign each to the right project before the team standup at 10am.
+**Q4 — Today:** Slack screenshots of the project list to each new hire; have them sign up individually; cross-check who landed where.
+**Q5 — Moment of truth:** The first time they see all three names in the team list with the right project tags, under 5 minutes.
+**Q6 — Failure mode:** Can't tell whether the invite arrived; the email looks like spam; they have to leave the page to find someone's email.
+**Q7 — Success in their words:** "It just worked. I sent the invites, they signed up, they were on the right project."
+**Q8 — Next thing:** Open the project where the first teammate landed and verify access.
+
+## Scenario 2: Returning admin adds a fourth member months later
+
+(…)
+
+## Coverage check
+
+| PRD story | Scenarios |
+|---|---|
+| E01 (Sign up) | S1 |
+| E02 (Invite) | S1, S2 |
+| E03 (Switch project) | S2 |
+| E04 (Dashboard) | S1 |
 ```
 
+## Anti-patterns Mantis rejects
+
+- **Personas instead of scenarios.** Mantis cares what people *do*, not who they are in the abstract.
+- **Scenarios with no emotion or context.** "User wants to sign up." — that's a backbone story restated. The point is the *state*.
+- **JTBD written as features.** "Wants the dashboard." Wrong layer. Reword to the goal: "Wants to know if the team is healthy at a glance."
+- **Q5 ("moment of truth") that's actually a feature description.** It should be the user's perception, not the UI.
+- **More than 8 scenarios.** Two of them are duplicates. Find which.
+
 ## Hand-off
-"Mantis → Mantis": run `wize-ux-design` next.
+
+> Scenarios are in `.wize/planning/ux/ux-scenarios.md`. They cover E01–E04. Next: page specs in `.wize/planning/ux/ux-design/` (one per screen). Tony, when you read these, the moment-of-truth lines tell you which interactions are load-bearing.

package/src/method-skills/2-plan-workflows/wize-validate-prd/workflow.md

@@ -2,26 +2,120 @@
 code: wize-validate-prd
 name: Validate PRD
 phase: 2-plan
-owner: wize-agent-pm   # Maria Hill (with Pepper + Mantis + Fury input)
-status: stub
+owner: wize-agent-pm   # Maria Hill (with Pepper, Mantis, Fury input)
+status: ready
 ---
 
 # Validate PRD
 
-**Goal.** Make sure the PRD is ready to leave Plan and enter Solutioning. Catch ambiguity before architecture is wasted on it.
+**Goal.** Decide whether the PRD is ready to leave Plan and enter Solutioning. Catch ambiguity before architecture is wasted on it. Cheap step; saves multiples of itself downstream.
+
+Maria Hill chairs. Pepper, Mantis, Fury contribute. Output is either `validated: true` written to `prd.md` or a punch list routed back to authors.
+
+## When to run
+
+After `wize-create-prd` produces `status: ready-for-validation`. Run once per major PRD change.
 
 ## Inputs
+
 - `.wize/planning/prd.md`
+- `.wize/planning/brief.md`
+- `.wize/planning/ux/trigger-map.md`
+- `.wize/planning/research.md` (when present)
+- `.wize/planning/tech-vision.md` (when Fury already wrote one)
 
 ## Outputs
-- Inline edits + a validation log appended to `prd.md`.
 
-## Checklist
-- [ ] Every scope item has measurable ACs.
-- [ ] No "TBD" in goals or constraints.
-- [ ] Mantis confirms UX is addressable.
-- [ ] Fury approves stack-family direction (or has explicit reason not to yet).
-- [ ] No open questions with `blocker` priority remain.
+- In-place edits to `prd.md` (validators add a "Validation log" section).
+- If validated: `prd.md` frontmatter flips to `status: validated`.
+- If not: a punch list, each item routed back to the owning agent.
+
+## Checklist (Maria Hill runs this)
+
+### Completeness
+- [ ] Vision sentence present and matches the brief.
+- [ ] 3–5 goals; each measurable; each tied to a trigger-map row.
+- [ ] Scope `In` and `Out` both populated; `Out` items have reasons.
+- [ ] Backbone stories cover every In-scope item.
+- [ ] Every In-scope item has 3–7 ACs.
+- [ ] Constraints + Assumptions + Dependencies all named.
+- [ ] NFR pointer present.
+- [ ] Open questions: each has an owner + deadline.
+
+### Quality
+- [ ] Every AC is observable (testable by Hawkeye without ambiguity).
+- [ ] No AC contains the word "easy", "intuitive", "responsive" without a numeric criterion.
+- [ ] INVEST: every backbone story passes (independent, negotiable, valuable, estimable, small, testable).
+- [ ] No goal repeats a brief constraint as if it were a goal.
+
+### Cross-checks
+- [ ] **Pepper:** every goal anchors on a row of the trigger map.
+- [ ] **Mantis:** every screen implied by the ACs is addressable in UX (no AC requires a UI the team hasn't agreed exists).
+- [ ] **Fury:** the PRD's NFR pointer doesn't contradict the tech-vision.
+- [ ] **Hawkeye preview:** at least the gate strategy is mentioned (advisory vs enforcing) for visibility.
+
+### Risk
+- [ ] No open question is marked `blocker` AND has a deadline past the PRD validation date.
+- [ ] Any assumption that, if wrong, kills > 30% of the scope has a verification plan.
+
+## Decision
+
+After the checklist:
+
+- **Validated** → flip frontmatter to `status: validated`. Append a Validation log block to `prd.md` with date + each signatory.
+- **Concerns** → write specific fix-asks per agent. Each fix-ask is concrete (line ref + what to change).
+- **Reject** → rare. Use when the brief itself is wrong; loop back to Pepper.
+
+## Validation log template (appended to `prd.md`)
+
+```markdown
+## Validation log — YYYY-MM-DD
+
+**Status:** validated
+
+**Signatories**
+- Maria Hill (PM)  — concerns: none
+- Pepper Potts (Analyst) — concerns: trigger-map row 3 needs a stronger signal name, will update.
+- Mantis (UX) — concerns: none, UX work starts next.
+- Fury (Solution Strategy) — concerns: none. NFR pointer confirmed.
+
+**Notes**
+- Open question "Brazil-only beta?" was answered by sales lead — added to *Out of scope: global launch* with rationale.
+```
+
+## Punch-list template (when not validated)
+
+```markdown
+## Validation punch list — YYYY-MM-DD
+
+**Status:** changes required
+
+**By Maria Hill — author rewrites:**
+- `AC-02-3`: "loads fast" is not observable. Rewrite with numeric target tied to Fury's NFRs.
+- Backbone story E04 fails INVEST-S (too large). Split into 2 stories.
+
+**By Pepper:**
+- Goal 2 is not anchored on a trigger-map row. Add row reference or remove.
+
+**By Mantis:**
+- AC-01-4 implies a "preview" screen that isn't in any UX list. Either spec the screen or cut the AC.
+
+**By Fury:**
+- NFR pointer cites old version of nfr-principles.md. Update.
+```
+
+## Anti-patterns
+
+- "Looks good to me" with no checklist run. The validation only counts if checked.
+- Validating to keep the schedule, knowing concerns are open. Document the concern.
+- Validating with no Hawkeye preview at all. Even a one-line gate strategy reference matters.
+
+## Hand-off
+
+If validated:
+
+> PRD validated. Mantis, you're up: `wize-ux-scenarios`. Fury, write or refresh `tech-vision.md` so Tony can read both.
+
+If not:
 
-## Outcome
-Either `validated: true` (timestamp + signatories) or list of fix-asks routed back to the right agent.
+> PRD not yet validated. Punch list lives at the bottom of `prd.md`. {{Author}} fixes by {{date}}; we re-run validation.

package/src/method-skills/3-solutioning/wize-check-implementation-readiness/workflow.md

@@ -2,21 +2,111 @@
 code: wize-check-implementation-readiness
 name: Check Implementation Readiness
 phase: 3-solutioning
-owner: wize-agent-architect   # Tony Stark (with Hawkeye)
-status: stub
+owner: wize-agent-architect   # Tony Stark (with Hawkeye + Hill)
+status: ready
 ---
 
 # Check Implementation Readiness
 
-**Goal.** Before any story enters dev, confirm everything is in place.
+**Goal.** Before any story enters dev, confirm everything is in place. This is the **gate at the end of Phase 3**. Cheap; saves multiples of itself downstream.
 
-## Checklist
-- [ ] Architecture doc exists and is signed (Tony + Fury approved).
-- [ ] All in-scope screens have UX specs (Mantis).
-- [ ] Design system tokens exist (Mantis).
-- [ ] Stories are sliced with ACs (Tony).
-- [ ] Hawkeye risk profile exists.
-- [ ] `.wize/config/tea.toml` policy choice is recorded.
+Tony chairs. Hill and Hawkeye sign off. Pepper and Mantis answer if pulled in.
+
+## Inputs
+
+- `.wize/planning/prd.md` (validated)
+- `.wize/planning/ux/ux-design/` + `index.md`
+- `.wize/planning/tech-vision.md`
+- `.wize/planning/nfr-principles.md`
+- `.wize/solutioning/architecture.md`
+- `.wize/solutioning/adrs/`
+- `.wize/solutioning/design-system/`
+- `.wize/solutioning/epics/` + `stories/`
+- `.wize/implementation/tea/risk-profile.md` (Hawkeye)
+- `.wize/config/tea.toml` (policy choice)
+
+## Outputs
+
+- `.wize/solutioning/readiness-{YYYY-MM-DD}.md` — checklist result + signoffs.
+
+## Checklist (Tony chairs)
+
+### Planning artifacts
+- [ ] PRD `status: validated`.
+- [ ] All open questions in PRD resolved (no `blocker` open).
+- [ ] Trigger map covers every PRD goal.
+
+### UX
+- [ ] Every In-scope item has at least one screen spec.
+- [ ] UX design index maps every screen → scenario → AC.
+- [ ] Design system tokens + components needed by the UX exist.
+
+### Strategy
+- [ ] Tech vision `status: aligned`.
+- [ ] NFR principles `status: aligned`; verifiers named for every non-negotiable.
+
+### Architecture
+- [ ] Architecture doc `status: ready-for-stories`.
+- [ ] ADRs cover every meaningful trade-off.
+- [ ] NFR check section answers *how* for each non-negotiable.
+
+### Stories
+- [ ] Every epic has 3–10 stories.
+- [ ] Every story has AC IDs from the PRD; the union per epic equals the epic's AC set.
+- [ ] No story is XL.
+- [ ] Each story names touch-points + `testid` + reuse of design-system components.
+
+### TEA
+- [ ] `tea-risk.md` exists.
+- [ ] `.wize/config/tea.toml` policy is committed (advisory / enforcing).
+- [ ] First-story `tea-design.md` is drafted (proof Hawkeye's contract works).
+
+### Cross-cutting
+- [ ] CI runs tests + validators on every PR.
+- [ ] Lint/format on commit.
+- [ ] Branch protection on `main`.
+- [ ] Secrets vault wired; no secret in repo.
 
 ## Outcome
-Either `ready: true` or list of fix-asks.
+
+- **Ready** → write the readiness file, link from `sprint-planning`.
+- **Concerns** → list specifically what's missing, owner per item, deadline. Re-check.
+
+## Output template
+
+```markdown
+---
+status: ready | concerns
+date: YYYY-MM-DD
+chair: Tony Stark
+signoffs:
+  - Maria Hill (PM)
+  - Hawkeye (TEA)
+  - Mantis (UX) — async
+  - Fury (Strategy) — async
+---
+
+# Implementation Readiness — {{project_name}}
+
+## Result: Ready
+
+## Notes
+- Architecture covers 5 epics and 28 stories.
+- TEA policy: advisory; NFR gate per epic.
+- 2 ADRs accepted in this gate review (ADR-008, ADR-009).
+
+## Open items (carry forward, not blockers)
+- Marketing site for launch — separate track, owned by …
+- Stripe-Atlas back-office decision — by date X.
+```
+
+## Anti-patterns
+
+- "Looks ready" without running the checklist.
+- Signing off before Hawkeye has risk-profiled.
+- Closing concerns without re-checking.
+- Leaving CI configuration "for later" — never gets done.
+
+## Hand-off
+
+> Implementation readiness signed off. Hill, kick `wize-sprint-planning`. Hawkeye, your gate cadence is set: design per story, trace + review + gate per story, NFR per epic.

package/src/method-skills/3-solutioning/wize-create-architecture/workflow.md

@@ -3,62 +3,230 @@ code: wize-create-architecture
 name: Create Architecture
 phase: 3-solutioning
 owner: wize-agent-architect   # Tony Stark
-status: stub
+status: ready
 ---
 
 # Create Architecture
 
-**Goal.** Design the system inside Fury's frame. Components, sequences, data flows, ADRs. Concrete enough that Shuri can implement and Hawkeye can test.
+**Goal.** Design the system inside Fury's frame. Components, sequences, data flows, ADRs. Concrete enough that Shuri can implement and Hawkeye can write `tea-design.md` for every story.
+
+Tony drives. Output lands in `.wize/solutioning/architecture.md` + `.wize/solutioning/adrs/`.
 
 ## Inputs
-- `.wize/planning/prd.md`
-- `.wize/planning/ux/ux-design/`
-- `.wize/planning/tech-vision.md` (Fury)
-- `.wize/planning/nfr-principles.md` (Fury)
-- `.wize/solutioning/design-system/` (Mantis)
-- `.wize/knowledge/document-project/` (if brownfield)
+
+- `.wize/planning/prd.md` (validated)
+- `.wize/planning/ux/ux-design/` (every architectural decision should make at least one screen possible)
+- `.wize/planning/tech-vision.md` (the frame)
+- `.wize/planning/nfr-principles.md` (the budget)
+- `.wize/solutioning/design-system/` (Mantis' tokens, when available)
+- Stack catalogs (overlays): `web-overlay/stack-catalog.md`, `app-overlay/stack-catalog.md`
+- `.wize/knowledge/document-project/` (brownfield only)
 
 ## Outputs
+
 - `.wize/solutioning/architecture.md`
-- `.wize/solutioning/adrs/{ADR-NNN}-{slug}.md`
+- `.wize/solutioning/adrs/ADR-NNN-{slug}.md` (one ADR per meaningful trade-off)
 
 ## Steps
-1. **Stack interview.** Tony asks the user (via Wizer) about preferences: language, runtime, deployment target. Consults overlay stack catalogs (web/app) if active. Records decisions.
-2. **Components.** List components, responsibilities, boundaries.
-3. **Sequences.** For each critical scenario from Mantis, write a sequence (text or PlantUML).
-4. **Data model.** Tables / collections / contracts.
-5. **Cross-cutting.** Auth, logging, observability, error handling.
-6. **ADRs.** Every significant trade-off becomes its own ADR: context, options, decision, consequences.
-7. **Validate against NFRs.** Check each Fury principle.
-8. **Hand-off to epics.**
+
+### 1. Stack interview (Tony asks; Wizer relays)
+
+Resolve every "TBD" the tech-vision left for Tony. Walk the stack catalog (active overlay) and decide, in order:
+
+- Language(s) + runtime(s).
+- Front-end framework + state lib + form lib.
+- Back-end framework or BaaS.
+- DB + ORM/query builder.
+- Auth.
+- Hosting + CI/CD.
+- Observability stack.
+- Test stack (links to `playwright-vitest.md` or `detox-maestro.md`).
+
+Decisions Tony makes silently are ADR candidates; decisions Fury already fixed don't get their own ADR.
+
+### 2. Components
+
+List components with one-line responsibility each. Boundaries before internals. Examples (web SaaS):
+
+| Component | Responsibility | Boundary |
+|---|---|---|
+| `web` | Server-rendered fullstack app (Next.js) | HTTPS to clients; SQL to db; HTTPS to auth-provider |
+| `db` | Source of truth for users, teams, billing (Postgres) | SQL only via PgBouncer |
+| `auth` | Identity provider (Supabase Auth) | OIDC to `web` |
+| `mailer` | Outbound transactional email | HTTPS to Resend |
+| `worker` | Outbox processor + scheduled jobs (pg_cron) | SQL to db; HTTPS to external APIs |
+
+### 3. Sequences (the critical ones)
+
+For each "moment of truth" in `.wize/planning/ux/ux-scenarios.md`, draw a sequence. Mermaid is fine; ASCII is fine.
+
+```mermaid
+sequenceDiagram
+  participant U as User
+  participant W as web
+  participant A as auth
+  participant D as db
+  U->>W: POST /signup
+  W->>A: signUp(email, password)
+  A-->>W: { user_id, session }
+  W->>D: INSERT user_id INTO accounts
+  D-->>W: ok
+  W-->>U: 302 /onboarding (sets cookie)
+  Note over W,U: total p95 ≤ 1s (NFR 1.A)
+```
+
+Annotate each sequence with the NFR target it must hit.
+
+### 4. Data model
+
+For every entity:
+
+- Name, columns, types, indexes.
+- Foreign keys + cascade behavior.
+- RLS policies if the stack supports them (Supabase, etc.).
+- Soft-delete vs hard-delete.
+
+Include a mini ERD (Mermaid `erDiagram`).
+
+### 5. Cross-cutting concerns
+
+For each, name the pattern and the library/component:
+
+- **Auth & session** — token shape, refresh, multi-device.
+- **Errors** — error class hierarchy, mapping to HTTP, user-facing copy.
+- **Logging** — structured (JSON), correlation IDs, sampling.
+- **Observability** — metrics emitter, traces, dashboards.
+- **Config** — env vars, secrets, feature flags.
+- **Background jobs** — outbox / queue / scheduler.
+- **Idempotency** — keys on write endpoints.
+- **i18n** — string source, translation pipeline.
+- **A11y** — token + library choices that uphold WCAG.
+
+### 6. NFR check (every category)
+
+Walk Fury's NFRs. For each non-negotiable, write *how* the architecture achieves it.
+
+- Perf: LCP ≤ 2.5s → edge runtime + RSC + image policy.
+- Security: PII in EU → DB in `eu-central-1`; backups in same region.
+- Reliability: 99.9% → single-region with multi-AZ; failover playbook in `adrs/ADR-007-failover.md`.
+- A11y: WCAG AA → Radix primitives + axe in CI.
+
+### 7. ADRs
+
+One ADR per meaningful trade-off. Format below. Number sequentially. Don't gold-plate; an ADR is a few paragraphs.
+
+### 8. Hand off
+
+Mark `architecture.md` `status: ready-for-stories`. Tony continues with `wize-create-epics-and-stories`.
 
 ## Architecture doc template
 
 ```markdown
+---
+status: ready-for-stories
+owner: Tony Stark
+created: YYYY-MM-DD
+---
+
 # Architecture — {{project_name}}
 
+## Summary
+{{One paragraph: stack family, runtime, primary data store, deploy target. The frame.}}
+
 ## Stack
-- Language: …
-- Runtime: …
-- Storage: …
-- Deploy: …
+- Language: TypeScript
+- Front-end: Next.js (App Router, RSC, edge runtime)
+- Back-end: Server Actions + Route Handlers
+- DB: Supabase Postgres + Drizzle ORM
+- Auth: Supabase Auth
+- Hosting: Vercel
+- Observability: Vercel + PostHog
+- Test: Vitest + Playwright (see playbook)
 
 ## Components
 | Component | Responsibility | Boundary |
 |---|---|---|
 
 ## Data model
-…
+- `users` (id PK, email UNIQUE, created_at)
+- `teams` (id PK, name, owner_id FK users)
+- `memberships` (user_id, team_id, role)
+- RLS: `auth.uid() = user_id` on all user-scoped tables.
+
+```mermaid
+erDiagram
+  USERS ||--o{ MEMBERSHIPS : has
+  TEAMS ||--o{ MEMBERSHIPS : has
+```
+
+## Sequences
+
+### S1: Sign-up
+{{sequence diagram + NFR annotation}}
+
+### S2: Invite teammate
+{{sequence diagram}}
 
 ## Cross-cutting
-- Auth: …
-- Observability: …
+- Auth & session: …
 - Errors: …
+- Logging: …
+- Observability: …
+- Config: …
+- Background jobs: …
+- Idempotency: …
+- i18n: …
+- A11y: …
 
 ## NFR check
-- Performance: …
-- Security: …
-- Reliability: …
-- Accessibility: …
-- Cost: …
+- Perf (1.A): how
+- Security (2.A): how
+- Reliability (3.A): how
+- Maintainability (4.A): how
+- A11y (5.A): how
+- Cost (6.A): how
+
+## ADRs
+See `.wize/solutioning/adrs/`.
 ```
+
+## ADR template
+
+```markdown
+---
+status: accepted | superseded | deprecated
+date: YYYY-MM-DD
+deciders: Tony, Fury
+supersedes: ADR-XXX
+---
+
+# ADR-007: {{slug}}
+
+## Context
+{{2–4 sentences: what forced the decision, what constraint is binding.}}
+
+## Options
+1. {{Option A}} — pros / cons / cost
+2. {{Option B}} — pros / cons / cost
+3. {{Option C}} — pros / cons / cost
+
+## Decision
+{{The pick. One sentence.}}
+
+## Consequences
+- **Now:** what we gain, what we accept.
+- **Later:** what we'll likely revisit and when.
+- **Related ADRs:** ADR-005, ADR-009.
+```
+
+## Anti-patterns Tony rejects
+
+- **Architecture without sequences.** A diagram with boxes is half a doc.
+- **NFR check left as "TBD".** Each non-negotiable answers *how*.
+- **ADRs for trivial choices** (which CSS file name) — saves nothing, costs trust.
+- **No ADR for genuinely contested choices** (auth provider, DB selection) — future-readers will re-litigate.
+- **Diagrams in proprietary format only.** Mermaid/ASCII version always present in markdown.
+
+## Hand-off
+
+> Architecture and 6 ADRs at `.wize/solutioning/`. Sequences hit the NFR targets. Hawkeye, you can write `tea-risk.md` against this. Tony continues with `wize-create-epics-and-stories`.