code-ai-installer 2.1.0 → 2.3.0

package/domains/product/locales/en/.agents/workflows/shape-prioritize.md

@@ -0,0 +1,263 @@
+# /shape-prioritize — Full Pipeline B (Build-vs-Cut Adversarial)
+
+> **Use when strategy is approved and the scope discussion is open.**
+> 6 sessions with adversarial on the **Build vs Cut axis**. Mediator synthesizes the final MoSCoW. Output — PRD (PDF).
+
+## When to Use
+
+- Strategy / vision / roadmap theme has already been agreed upon (within the team or in a previous /ship-right-thing session)
+- Team disagrees: "build everything" vs "cut ruthlessly"
+- Tight timeline, prioritization is important
+- A final PRD with justified scope and a scope-rationale appendix is needed
+- High-stakes release (quarter-defining initiative)
+
+## When NOT to Use
+
+- Strategy not approved → `/ship-right-thing`
+- Scope approved, only a PRD is needed → `/spec`
+- Quick assessment → `/quick-pm`
+- Solo-PM scope decision without disagreement → `/spec` (with PM acting in consultation)
+
+## Decision Tree (clarifying questions in COND-01)
+
+```
+Strategy approved?
+  ├── NO → /ship-right-thing (strategy synthesis needed first)
+  └── YES → continue
+
+Scope debate open?
+  ├── YES (team arguing / timeline pressure) → /shape-prioritize (this workflow)
+  └── NO (scope known) → /spec
+
+Risk of cut (cutting something valuable) or risk of bloat (overcommit) — equally high?
+  ├── YES → /shape-prioritize (adversarial is valuable)
+  └── NO (one side is obvious) → /spec
+
+Is exec-review presentation with scope-rationale needed?
+  ├── YES → /shape-prioritize (mini deck part of output)
+  └── NO → /shape-prioritize still (PRD is primary), mini deck optional
+```
+
+## Pipeline (6 sessions)
+
+```
+Session 1: CONDUCTOR → DISCOVERY (scope intake)                         → session-1-handoff.md
+Session 2: CONDUCTOR → BUILD-CAMP (pm α + tech_lead α)                  → session-2-handoff.md
+Session 3: CONDUCTOR → CUT-CAMP (pm β + tech_lead β)                    → session-3-handoff.md
+Session 4: CONDUCTOR → MEDIATOR (Scope Synthesis)                       → session-4-handoff.md
+Session 5: CONDUCTOR → PM (final PRD) → UX_DESIGNER + DATA_ANALYST      → session-5-handoff.md
+Session 6: CONDUCTOR → DESIGNER → LAYOUTER → RELEASE_GATE               → PDF
+```
+
+Each session starts from scratch. Unlike /ship-right-thing, PM α and β work **paired with Tech Lead α and β** — two-agent camps (parallel expertise).
+
+## Input
+
+| Field | Required | Description |
+|-------|:--------:|-------------|
+| Approved strategy / roadmap theme | ✅ | Where we pull from: which initiative |
+| Vision / NSM / OKR | ✅ | From the approved strategy |
+| Time box (release horizon) | ✅ | Quarter / release date |
+| Team composition | ✅ | Senior / mid / junior count, velocity baseline |
+| Tech constraints | ✅ | Stack, integrations, existing tech debt |
+| Compliance constraints | ⬚ | SOC 2 / GDPR / HIPAA / industry-specific |
+| Initial backlog (if available) | ⬚ | What is already being discussed in the team |
+| Budget ceiling | ⬚ | For Tech Lead feasibility with budget constraint |
+
+## Gates and Deliverables
+
+### Session 1 — Scope Intake Gate
+
+- **COND-01** (Conductor): Interview Brief
+  - 5+ clarifying questions (current scope debate, team composition, constraints)
+  - Strategy Brief loaded (pointer or inline)
+  - Scope-debate context documented
+
+- **DISC-01** (Discovery): Scope-focused Discovery Brief
+  - Which JTBD are in scope (within the approved strategy) — and which are NOT in scope
+  - Assumptions: what we consider true for the scope decision
+  - Open questions (that require experiments / pilots)
+  - Focus on end-user JTBD (buyer has already justified the strategy)
+
+### Session 2 — Build-Camp (Camp Alpha)
+
+- **PM-01α** (PM α, paired with Tech Lead α): full Scope Proposal
+  - RICE-scored backlog (top 15-20 items)
+  - **Each item with rationale** "why it is mandatory now"
+  - Include-bias: every idea is presumed necessary, burden of proof — show it can be deferred
+  - Kano balance (may include delighters as argument "we won't win without this")
+  - Out-of-scope: minimal, explicit
+
+- **TECH-01α** (Tech Lead α, paired with PM α): Build Feasibility
+  - "Yes, we will deliver in N team-weeks with [team setup]"
+  - T-shirt by the **lower bound** with parallelization rationale
+  - Existing-stack leverage points
+  - Allowed trade-offs (but not cuts)
+  - Managed risk posture: every risk with mitigation
+
+### Session 3 — Cut-Camp (Camp Beta)
+
+- **PM-02β** (PM β, paired with Tech Lead β): Minimum Scope Proposal
+  - **MoSCoW**: Must — only what we cannot launch without
+  - Should / Could / Won't (this release) with rationale per cut
+  - Exclude-bias: every idea is presumed unnecessary, burden of proof — why it is mandatory now and not in Q2
+  - Validation paths for deferred items (experiments in Q2)
+  - Out-of-scope: as explicit as possible, with rationale
+
+- **TECH-02β** (Tech Lead β, paired with PM β): Hidden Risks + Cut Rationale
+  - T-shirt by the **upper bound** with hidden dependency rationale
+  - Hidden integrations, migration costs
+  - Technical debt implications (what cannot start without prior work)
+  - Compliance phantoms (SOC 2 / GDPR riding tag)
+  - 3rd party SLA risks
+  - "What cannot start without [X]" explicitly for each deferred item
+
+### Session 4 — Mediator Scope Synthesis
+
+- **MED-01** (Mediator): Unified Scope Decision
+  - **Independence check** (contamination: true/false)
+  - Evidence audit per camp (coverage %, strength per position)
+  - Position map **per item** (Alpha Build / Beta Cut / overlap / disagreement)
+  - Disagreement analysis per item (Factual / Value / Risk / Methodological)
+  - Scoring (5 dimensions): Evidence / Coherence / Delivery Feasibility / Risk / Value Density
+  - Strengths & Weaknesses per camp
+  - **Synthesis Path**: Adopt Build / Adopt Cut / Hybrid (often hybrid — Must from both + Should from one)
+  - **Final MoSCoW** — source-pointer per item
+  - Tradeoff map (what was cut and why)
+  - Delivery estimate (team-weeks)
+
+### Session 5 — Spec (Final PRD for the approved scope)
+
+- **PM-SPEC** (PM, neutral — executes Mediator synthesis):
+  - Final PRD (10 sections)
+  - User stories INVEST + Gherkin AC for Mediator-approved items
+  - Epic breakdown
+  - Scope rationale appendix (from Mediator Tradeoff Map)
+  - NFR (from Mediator synthesis + Tech reviews)
+  - Success metrics (from approved Strategy)
+  - Rollout plan (pilot → beta → GA with criteria)
+
+- **UX-01** (UX Designer, in parallel):
+  - User flows for Mediator-approved scope
+  - Low-fi wireframes for Must items
+  - Accessibility checklist (WCAG AA)
+  - Design brief for Designer
+
+- **DATA-01** (Data Analyst, in parallel):
+  - Metric plan for approved scope
+  - Validation experiments for deferred items (Q2 planning)
+  - SaaS Impact Model (cut items — potential impact as reference)
+  - Instrumentation requirements
+
+Note: in Full B Session 5, **Tech Lead** is not required to be active — feasibility was already built in Sessions 2-3 and synthesized by Mediator. If PRD spec reveals new tech concerns — Reverse Handoff to Tech Lead.
+
+### Session 6 — Document
+
+- **DS-01** (Designer): Design Spec
+  - PRD document IA with scope rationale appendix
+  - Mini deck (10 slides): scope decision rationale for stakeholder review (not a full exec deck — Full B focus is on PRD)
+  - Visualizations: MoSCoW matrix, Tradeoff map, Epic breakdown, Risk heatmap
+
+- **LY-01** (Layouter): HTML → PDF
+  - PRD HTML (25-30 pages + appendix)
+  - Mini deck HTML (10 slides, if required)
+  - Self-contained, print-ready
+
+- **RG-01** (Conductor): Release Gate sign-off
+
+## Build-Camp vs Cut-Camp Rules
+
+**Build-Camp presumption:** every idea on the table is needed. Burden of proof — show it can be deferred.
+- T-shirt by the lower bound.
+- Parallelization through team setup.
+- Existing-stack leverage.
+- Managed risk (mitigation per risk).
+- Output tone: "here is the path — this is achievable".
+
+**Cut-Camp presumption:** every idea is unnecessary. Burden of proof — show it is mandatory now.
+- T-shirt by the upper bound.
+- Hidden dependencies surfaced.
+- Compliance / tech debt / 3rd party risks explicit.
+- Validation paths for deferred items.
+- Output tone: "here are the risks — here is what we defer, and here is how we validate".
+
+Both use the **same** Discovery Brief and Strategy Brief as input.
+
+**Camp discipline self-check (before handoff):**
+- Build-Camp: "Did I cut an item that was in Discovery?" → if yes, reconsider as Include.
+- Cut-Camp: "Did I include an item on a 'nice-to-have' rationale?" → if yes, move to Should / Could.
+
+## Adversarial Independence (pipeline rule 4)
+
+Build-Camp (PM α + Tech Lead α) and Cut-Camp (PM β + Tech Lead β) work independently.
+
+- They do not see each other until Mediator.
+- Paired agents (PM + Tech Lead) within the same camp coordinate through Conductor handoff in the same session.
+- Beta receives the Alpha deliverable read-only via camp filter in Session 3.
+- Contamination check by Mediator — same as in Full A.
+
+## Severity levels
+
+- **P0 Blocker:** absence of one of the camps, scope without justifications, contamination critical, Mediator without synthesis
+- **P1 Gap:** thin evidence in Cut-camp (acceptable, Mediator will audit); weak rationale per item
+- **P2 Note:** mini deck for non-exec stakeholders; edge-case items
+
+## Escalation paths
+
+- **Stalemate with equal evidence** — both camps are strong, delta < 0.5, Mediator cannot choose a path → escalate to user with two scope options and recommended
+- **Resource block** — scope > capacity even in Cut-Camp minimum → pause, re-scope within constraints (return to Strategy)
+- **Compliance block** — Cut-Camp surfaces a compliance requirement that is critical for release → pause, legal review
+- **Hidden strategy gap** — revealed that a strategy assumption does not hold → pause, return to /ship-right-thing mini-session
+
+## Health metrics for this workflow
+
+| Metric | Healthy | Problematic | Action |
+|--------|:-------:|:-----------:|--------|
+| Build-Cut divergence | Delta 10-30% items | Delta > 50% | Possible misalignment — Mediator investigate |
+| Cut rationale depth | Per item with tech + user evidence | Hand-wave "nice to cut" | Reverse Handoff to Cut-Camp |
+| Build feasibility confidence | ≥ 70% items with Spike or proven pattern | < 40% | Reverse Handoff to Build-Camp + Tech Lead |
+| Final MoSCoW Must % | 30-50% total items | > 70% Must | Cut-Camp underdelivered, Mediator re-audit |
+| Delivery estimate vs time box | ≤ 100% capacity | > 120% | Re-scope in Mediator |
+
+## Output
+
+- `prd.pdf` — final PRD with justified scope (25-35 pages + appendix)
+- `prd.pdf` → Appendix: Scope Rationale (from Mediator Tradeoff Map)
+- Optional `scope-decision-deck.pdf` — 10-slide mini deck for stakeholder review
+- All `session-N-handoff.md` in `docs/product/` (N = 1..5)
+- `$board` history with metrics
+
+## Example — TeamFlow AI 1:1: Shape MVP scope (Full B)
+
+**Request:** "Strategy approved (AI summaries). How much do we put in MVP: summary only / + coaching / + admin dashboard? Timeline Q1."
+
+**Mode:** /shape-prioritize (strategy approved, scope debate open, 3 MVP scenarios)
+
+**Session 1 (Discovery scope intake):**
+- Which JTBD are in scope Q1: end-user summary (Must-have JTBD from previous discovery). Buyer dashboard — under review.
+- Assumptions: managers actually read summaries (validated in Q4 pilot ✅). CPO pay premium for dashboard (⚠️ assumed).
+
+**Session 2 (Build-Camp PM α + TL α):**
+- RICE backlog top 12 items. Include: summary + coaching + dashboard + export + RU + Slack.
+- TL: 12 team-weeks with 3 senior + 2 mid, parallelization. Existing stack covers 80%.
+
+**Session 3 (Cut-Camp PM β + TL β):**
+- MoSCoW: Must = summary + rating feedback. Should = coaching (validate first). Could = dashboard (assumption unvalidated). Won't = RU, Slack, export.
+- TL β: Dashboard revealed dependency on new aggregation service (+3 team-weeks). Compliance on OpenAI TOS in-progress (P1 gap).
+
+**Session 4 (Mediator Scope Synthesis):**
+- Scoring: Build Total 7.0, Cut Total 7.6 (delta 0.6 → slight Cut advantage).
+- Synthesis: **Hybrid** — Must = summary + rating (from both), Should = coaching (with validation gate at W4), Could = dashboard (spike in W2 for CPO interviews), Won't = RU / Slack / export.
+- Tradeoff map: dashboard deferred until validation (P1 — will disconnect 20% premium revenue hypothesis), coaching Should with go/no-go gate.
+
+**Session 5 (Spec):**
+- PRD: 8 user stories for Must + coaching, 3 epics, rollout pilot.
+- UX: primary flow (generate summary), coaching prompt panel (Should, MVP-2).
+- Data: NSM tree, 3 hypotheses, A/B design for coaching.
+
+**Session 6:**
+- PRD 28 pages + scope rationale appendix (4 pages).
+- Mini deck 10 slides (scope decision rationale for VP Eng review).
+- Release Gate sign-off.
+
+**Result:** 6 sessions × ~1.5h = 9h for unified scope with explicit tradeoffs, validated through adversarial debate.

package/domains/product/locales/en/.agents/workflows/ship-right-thing.md

@@ -0,0 +1,242 @@
+# /ship-right-thing — Full Pipeline A (Customer-vs-Business Adversarial)
+
+> **Use when strategy is not approved and a full plan from discovery to PRD + PDF is needed.**
+> 6 sessions with adversarial on the **Customer vs Business axis**. Mediator synthesizes. Output — PRD + Product Review Deck (PDF).
+
+## When to Use
+
+- New product direction, pivot, large initiative
+- High stakes: decision will affect quarters of work
+- A second perspective on strategy is needed (Customer vs Business)
+- Final artifact — Product Review Deck for exec sign-off
+- Strategy alternatives are not obvious — adversarial debate is needed, not solo-planning
+
+## When NOT to Use
+
+- Small fix → `/quick-pm`
+- Strategy approved, scope under discussion → `/shape-prioritize`
+- Scope approved → `/spec`
+- Known roadmap theme, only a detailed PRD is needed → `/spec`
+
+## Decision Tree (clarifying questions in COND-01)
+
+```
+Strategy / vision / NSM approved?
+  ├── NO → /ship-right-thing (this workflow)
+  └── YES → /shape-prioritize or /spec
+
+Is a PDF needed for exec-review?
+  ├── YES → /ship-right-thing (Product Review Deck — Full A output)
+  └── NO + only PRD → /spec or /shape-prioritize (if scope is open)
+
+Stakes:
+  ├── High (quarters of work, $100K+ investment) → /ship-right-thing
+  └── Low (reversible, weekly decision) → /quick-pm
+```
+
+## Pipeline (6 sessions)
+
+```
+Session 1: CONDUCTOR → DISCOVERY                                        → session-1-handoff.md
+Session 2: CONDUCTOR → CUSTOMER-CHAMPION (product_strategist α)         → session-2-handoff.md
+Session 3: CONDUCTOR → BUSINESS-CHAMPION (product_strategist β)         → session-3-handoff.md
+Session 4: CONDUCTOR → MEDIATOR (Strategy Synthesis)                    → session-4-handoff.md
+Session 5: CONDUCTOR → PM → UX_DESIGNER + TECH_LEAD + DATA_ANALYST      → session-5-handoff.md
+Session 6: CONDUCTOR → DESIGNER → LAYOUTER → RELEASE_GATE               → PDF
+```
+
+Each session starts from scratch — the handoff file is the only source of context.
+
+## Input
+
+| Field | Required | Description |
+|-------|:--------:|-------------|
+| Product question | ✅ | What we are solving, for whom — in 1 sentence |
+| Segment | ✅ | SMB / mid-market / enterprise |
+| Current metrics | ✅ | ARR, churn, NRR, LTV/CAC, Payback |
+| Constraints | ✅ | Timeline, team, tech, compliance, budget |
+| Planning horizon | ⬚ | Usually 1-2 quarters for Q1 delivery |
+| Existing product context | ⬚ | Current features / stack / recent incidents |
+| Competitive landscape | ⬚ | Key competitors (Mediator uses in evidence audit) |
+
+## Gates and Deliverables
+
+### Session 1 — Discovery Gate
+
+- **COND-01** (Conductor): Interview Brief
+  - 5+ clarifying questions answered
+  - Pipeline mode confirmed
+  - Scope and anti-scope defined
+  - `$board` created with the full list of gates
+
+- **DISC-01** (Discovery): Discovery Brief
+  - JTBD (functional / emotional / social) — **separately for buyer and end-user** (B2B critical)
+  - Problem statements (top 3-5) — formula [actor][context][pain][root cause][evidence]
+  - Assumption map (4 quadrants: value / usability / feasibility / viability) with top risky
+  - Evidence inventory with quality markers (✅ Verified / ⚠️ Old / 🔮 Assumed)
+  - Open questions for Session 2
+
+### Session 2 — Customer-Champion (Camp Alpha)
+
+- **STRAT-01α** (product_strategist α): Strategy Brief Alpha
+  - Vision (from user outcomes — "managers save 2h/week...")
+  - NSM — user-value metric (WAM using feature × rating ≥ 4/5)
+  - OKR (1-3 Objectives + 3-5 KR each, with numbers and deadlines)
+  - Roadmap (theme-based: Now / Next / Later)
+  - Kano balance (must-have / performance / delighters with user-pain focus)
+  - Evidence traceability per position
+  - Top 5 risks with mitigation
+
+### Session 3 — Business-Champion (Camp Beta)
+
+- **STRAT-02β** (product_strategist β): Strategy Brief Beta
+  - Vision (from business outcomes — "+$1M ARR in HR-tech segment...")
+  - NSM — business-value metric (ARR per seat, NRR by segment)
+  - OKR (ARR, CAC payback, NRR, win rate focus)
+  - Roadmap (theme-based, opens sales motion / pricing / moat first)
+  - Rationale: revenue / margin / competitive moat
+  - Evidence traceability per position
+  - Top 5 risks with mitigation
+
+### Session 4 — Mediator Strategy Synthesis
+
+- **MED-01** (Mediator): Unified Strategy Brief
+  - **Independence check** (contamination: true/false)
+  - Evidence audit per camp (coverage %, strength per position)
+  - Position map (Alpha vs Beta on key themes: Vision / NSM / OKR / Roadmap / Risks)
+  - Disagreement analysis (Factual / Value / Risk / Methodological classification)
+  - Scoring (5 dimensions × 0-10): Evidence / Coherence / Feasibility / Risk / User-Business Fit
+  - Strengths & Weaknesses per camp
+  - **Synthesis Path**: Adopt Alpha / Adopt Beta / Hybrid (with explicit rationale)
+  - **Unified deliverable** (Vision, NSM, OKR, Roadmap) — source-pointer per element
+
+### Session 5 — Planning + Specification
+
+- **PM-SPEC** (PM): Final PRD + prioritized backlog
+  - PRD 10 sections (Overview → Rollout)
+  - RICE-scored backlog top 15
+  - User stories INVEST + Gherkin AC
+  - Epic breakdown 4-8 epics
+  - Dependencies map with UX / Tech / Data
+  - Out-of-scope explicit
+
+- **UX-01** (UX Designer, in parallel with Tech + Data):
+  - User flows (primary + secondary + edge)
+  - Low-fi wireframes
+  - Accessibility checklist (WCAG AA baseline)
+  - Design brief for Designer downstream
+
+- **TECH-01** (Tech Lead, in parallel):
+  - Architecture review
+  - Feasibility matrix (t-shirt per item with rationale)
+  - NFR (9 categories: performance / security / scalability / compliance / observability / integrations / accessibility / i18n / privacy)
+  - Risk Register (5-10 risks with P×I + mitigation + owner)
+  - Dependencies Map + Critical Path (team-weeks floor)
+  - Epic breakdown with critical path membership
+
+- **DATA-01** (Data Analyst, in parallel):
+  - Metric Plan (tree: NSM → inputs → guardrails → leading)
+  - Hypotheses formalized (3-7 with threshold + timeframe)
+  - Experiment designs (per primary hypothesis with sample size / MDE / guardrails)
+  - SaaS Impact Model (ARR / churn / NRR / LTV/CAC / Rule of 40)
+  - AARRR funnel (B2B-adapted: Activation ≠ signup)
+  - Instrumentation requirements (events / properties / cohorts / privacy)
+
+### Session 6 — Document Finalization
+
+- **DS-01** (Designer): Design Spec
+  - PRD document IA (10 sections)
+  - Product Review Deck structure (14-15 slides, Why→What→How→When→Measure→Risks→Ask arc)
+  - Visual hierarchy (typography, color B2B sober, spacing 8pt)
+  - Key visualizations spec (roadmap, metric tree, RICE, risk heatmap, flow summary)
+  - Layouter handoff spec (8 sections)
+
+- **LY-01** (Layouter): HTML → PDF
+  - Self-contained PRD HTML (semantic, print-ready)
+  - Self-contained Deck HTML (15-20 slides)
+  - Embedded CSS, page-break hints, @page rules
+  - TOC with anchors
+  - Visualizations: CSS Grid / SVG / Mermaid / Chart.js with fallback
+
+- **RG-01** (Conductor): Release Gate sign-off
+  - All gates [✓]
+  - Visual check PDF (page breaks, content intact)
+  - User sign-off received
+
+## Adversarial Independence (pipeline rule 4)
+
+**Critical:** Alpha and Beta work independently in Sessions 2 and 3.
+
+- Both receive the same Discovery Brief (from Session 1).
+- They do not share artifacts, they do not see the other camp's position.
+- Beta handoff in Session 3 is marked `team: beta` + contains Alpha deliverable read-only via camp filter.
+- Conductor applies camp filter when forming Session 3 handoff (Alpha Strategy → Beta read-only).
+
+**Contamination check** (Mediator, Step 1):
+- Direct quotes from Alpha in Beta without marking → contamination: true.
+- Structural mirroring (Beta vision = rephrased Alpha) → flag.
+- When contamination: `confidence` in synthesis is lowered, user notified, possible rerun of one of the camps.
+
+## Severity levels
+
+- **P0 Blocker:** absence of Discovery Brief, one of the camps, Mediator synthesis; contamination = true (critical case)
+- **P1 Gap:** weak evidence in one camp (acceptable, Mediator will account for it during scoring), evidence coverage 50-79%
+- **P2 Note:** UX wireframes for non-UI initiatives; low-priority disagreements
+
+## Escalation paths
+
+- **Stalemate** (Mediator cannot synthesize, both camps are weak) → escalate to user with two options + rationale + recommended (usually rerun one of the camps)
+- **Contradictory evidence** between camps → additional discovery mini-session (loop back to Discovery)
+- **Compliance block** found in Session 5 (Tech Lead review) → pause pipeline, legal / security review, return
+- **Resource block** (scope > capacity) → loop back to Mediator for re-synthesis with cut constraints or escalate budget
+- **User non-sign-off** on Mediator synthesis → Reverse Handoff to Discovery for more evidence
+
+## Health metrics for this workflow
+
+| Metric | Healthy | Problematic | Action |
+|--------|:-------:|:-----------:|--------|
+| Session 1-3 velocity | 1 session / 1-2 days | > 3 days / session | Check clarification quality |
+| Evidence coverage (per camp) | ≥ 80% | < 50% | Discovery loop |
+| Contamination | false | true | Rerun or partial rerun |
+| Mediator synthesis delta (Total) | < 1.5 (hybrid expected) | > 3.0 | Possible bias, re-audit |
+| Session 5 parallel completion | All 4 parallel agents ✅ | Sequential due to dependencies | Unblock by PM |
+
+## Output
+
+- `prd.pdf` — final PRD for the team (30-45 pages)
+- `product-review-deck.pdf` — for exec sign-off (15-20 slides)
+- All `session-N-handoff.md` in `docs/product/` (N = 1..5)
+- `$board` history with metrics
+
+## Example — TeamFlow AI 1:1 Summarization (end-to-end)
+
+**Request:** "Should we launch AI 1:1 summarization for managers? What should the strategy be?"
+
+**Mode:** /ship-right-thing (strategy not approved, high stakes — $M initiative)
+
+**Session 1 (Discovery):**
+- Conductor: 5+ questions (segment HR-tech, ARR $8M, buyer=CPO, end-user=Manager).
+- Discovery: 8 manager interviews (✅ Verified), CPO 2 interviews (⚠️ Old). 3 problems (top: 2h/week prep pain). 8 assumptions (top risky: "managers actually read summaries"). Coverage 72%.
+
+**Session 2 (Customer-Champion α):**
+- Strategy Alpha: Vision "managers save 2h/week", NSM = "40% MAM × 4.0 rating", OKR focus adoption + retention.
+
+**Session 3 (Business-Champion β):**
+- Strategy Beta: Vision "unlock $1M ARR HR-tech segment", NSM = "ARR per seat × seats-with-activity", OKR focus sales motion + ARPU.
+
+**Session 4 (Mediator):**
+- Scoring: Alpha Total 7.3, Beta Total 7.4 (delta < 0.5 → parity).
+- Synthesis: **Hybrid** — Alpha vision (user) + Beta risk lens (adoption stall) + hybrid OKR (40% MAM + 10 HR-tech logo wins).
+
+**Session 5 (Planning):**
+- PM PRD: 10 user stories, RICE-scored, 5 epics, rollout pilot→beta→GA.
+- UX: primary flow (generate summary post-1:1), accessibility WCAG AA.
+- Tech: 10 team-weeks critical path, 5 managed risks.
+- Data: NSM tree, 3 hypotheses, 1 A/B design (coaching prompts).
+
+**Session 6 (Document):**
+- Designer: PRD 35 pages + Deck 15 slides.
+- Layouter: HTML → PDF, self-contained.
+- Release Gate: user sign-off, PDF published.
+
+**Result:** 6 sessions × ~2h = 12h for a unified plan, validated with adversarial debate, evidence-based.