code-ai-installer 2.2.0 → 2.4.0

package/domains/product/locales/en/.agents/workflows/spec.md

@@ -0,0 +1,229 @@
+# /spec — PRD-only Pipeline (1 session, no adversarial)
+
+> **Use when strategy and scope are approved — a formal PRD is needed.**
+> 1 session, no adversarial. Output — PRD (markdown) or PRD PDF (optional).
+
+## When to Use
+
+- Strategy approved (in previous `/ship-right-thing` sessions or from an external process)
+- Scope approved (in `/shape-prioritize` or from a roadmap document)
+- Team is ready to start implementation — a PRD is needed for handoff
+- Feature / initiative within an approved roadmap theme
+- Team alignment already achieved — spec is needed for engineering / UX / QA
+
+## When NOT to Use
+
+- Scope is open → `/shape-prioritize`
+- Strategy unclear → `/ship-right-thing`
+- A quick assessment without a full PRD is needed → `/quick-pm`
+- Exploratory / discovery-heavy initiative → `/ship-right-thing`
+
+## Decision Tree (clarifying questions in COND-01)
+
+```
+Strategy approved?
+  ├── NO → /ship-right-thing
+  └── YES → continue
+
+Scope approved?
+  ├── NO → /shape-prioritize
+  └── YES → continue
+
+Success =
+  ├── Full PRD for engineering handoff → /spec (this workflow)
+  ├── Short brief for discussion → /quick-pm
+  └── Deep strategy rethink → /ship-right-thing
+```
+
+## Pipeline (1 session)
+
+```
+CONDUCTOR → PM → UX_DESIGNER + TECH_LEAD → PRD (markdown)
+             ↓
+            DATA_ANALYST (for success metrics validation)
+```
+
+Sequential-parallel:
+- PM launches PRD draft.
+- UX Designer + Tech Lead work in parallel on PM output.
+- Data Analyst — sanity check success metrics.
+- PM finalizes PRD, integrating UX + Tech + Data input.
+
+## Input
+
+| Field | Required | Description |
+|-------|:--------:|-------------|
+| Initiative name | ✅ | What we are specifying |
+| Approved strategy context | ✅ | NSM, OKR, roadmap theme (reference or inline) |
+| Approved scope | ✅ | Backlog / MoSCoW / decision memo |
+| User segments | ✅ | Buyer vs end-user (from Discovery / strategy) |
+| Constraints | ✅ | Timeline, team, NFR (from Tech expectations) |
+| Existing flows / screens | ⬚ | If extension of existing UI |
+| Success metrics expected | ⬚ | Usually inherited from approved strategy NSM |
+
+## Gates and Deliverables
+
+### COND-01 — Clarification + Scope Review
+
+- **Conductor**:
+  - 5+ clarifying questions (scope boundaries, timeline, NFR expectations, audience)
+  - Validate approved scope (sanity check — are there open questions?)
+  - Identify dependencies (other teams / external services / compliance)
+  - `$board` created with 5 gates (COND-01 / PM-SPEC / UX-01 / TECH-01 / LITE-RG)
+
+### PM-SPEC — Core PRD
+
+- **PM**: Final PRD via `$prd-template` (10 sections):
+  1. Overview (problem + audience + business context)
+  2. Goals & Success Metrics (NSM / OKR link)
+  3. Solution Approach + non-goals
+  4. User Stories (INVEST) — 5-10 stories
+  5. Acceptance Criteria (Gherkin) per story
+  6. NFR (compact)
+  7. Risks (top 5 with mitigation)
+  8. Rollout Plan (pilot → GA)
+  9. Dependencies
+  10. Open Questions
+
+- **PM** also:
+  - Epic breakdown (3-5 epics)
+  - Out-of-scope explicit
+
+### UX-01 — User Flows + Wireframes (in parallel with Tech, if UI)
+
+- **UX Designer** via `$user-flow` + `$design-brief`:
+  - User flows for primary + secondary flows
+  - Low-fi wireframes for key screens
+  - Accessibility checklist (WCAG AA baseline)
+  - Design brief for downstream visual designer (optional, if hi-fi is required)
+  - B2B persona delineation (buyer flows / end-user flows / shared)
+
+If initiative is non-UI (backend / infrastructure) — UX skip, marked as N/A on `$board`.
+
+### TECH-01 — Feasibility + NFR (in parallel with UX)
+
+- **Tech Lead** (neutral mode, no camp):
+  - Architecture review
+  - Feasibility check (Yes / Yes-caveat / No + rationale per item)
+  - NFR (compact — top 5-7 categories: performance / security / compliance / observability / integrations)
+  - Risk Register (top 3-5 with P×I + mitigation + owner)
+  - Integration map (dependencies on other services / teams)
+  - Epic breakdown (shallow, shared with PM)
+
+### DATA-01 — Metric Plan (optional if scope contains new metrics)
+
+- **Data Analyst**:
+  - Metric tree (NSM + top 3 inputs + guardrails)
+  - Hypothesis formalization (1-3, if applicable)
+  - Instrumentation requirements (events + properties)
+  - Baseline and target values
+
+If success metrics are inherited as-is from the approved strategy and instrumentation already exists — Data skip, marked N/A.
+
+### LITE-RG — PRD Finalization
+
+- **PM** merges UX + Tech + Data input
+- Self-review checklist (PRD Content Checklist — below)
+- Cross-check acceptance criteria for consistency with user stories
+- User sign-off
+
+### Optional: DS-01 + LY-01 (if PDF is needed)
+
+- Designer: compact design spec (10-15 page PRD layout).
+- Layouter: HTML → PDF.
+- Adds +1 session (Session 2 Spec becomes a Full A-style document session).
+
+## PRD Content Checklist (LITE-RG criteria)
+
+- [ ] Problem (clear, evidence-backed, from approved Discovery pointer)
+- [ ] Audience (buyer + end-user from Discovery)
+- [ ] Solution approach (1-2 paragraphs, no implementation details)
+- [ ] Success metrics (primary NSM + guardrails — from approved strategy)
+- [ ] User stories (INVEST — Independent / Negotiable / Valuable / Estimable / Small / Testable)
+- [ ] Acceptance criteria (Gherkin Given/When/Then)
+- [ ] Non-functional requirements (performance / security / compliance / accessibility)
+- [ ] Out-of-scope (explicit list)
+- [ ] Risks (top 3-5 with mitigation + owner)
+- [ ] Rollout plan (pilot / beta / GA with criteria + rollback)
+- [ ] Dependencies (other teams / external / compliance)
+- [ ] Open questions (owner + deadline)
+
+## Severity levels
+
+- **P0 Blocker:** PRD without success metrics; user stories without AC; NFR void; audience undefined
+- **P1 Gap:** Skipped UX for an initiative where UI is affected; missing rollout plan; thin dependencies
+- **P2 Note:** Optional PDF; expanded risks beyond top 5
+
+## Escalation paths
+
+- **Scope reveals strategy gap** (new contradiction with approved strategy) → pause, escalate to /ship-right-thing mini-session
+- **Scope reveals feasibility blocker** (Tech Lead surfaces hard-no) → pause, escalate to PM / stakeholder for re-scope
+- **Compliance block** found (SOC 2 / GDPR / HIPAA) → pause, legal review, return
+- **Timeline conflict** (scope > capacity) → escalation: cut scope or extend timeline (user decision)
+
+## Health metrics
+
+| Metric | Healthy | Problematic | Action |
+|--------|:-------:|:-----------:|--------|
+| PRD completeness | 10/12 checklist items | < 8 | Do not pass to engineering |
+| User stories AC coverage | 100% stories have AC | < 80% | Reverse Handoff PM |
+| NFR coverage | ≥ 5 categories | < 3 | Reverse Handoff Tech Lead |
+| Session duration | ≤ 3h focused | > 6h | Scope too big for Spec → consider split |
+
+## Output
+
+- `prd.md` (main deliverable) — 10-20 pages markdown
+- Optional `prd.pdf` (if Designer + Layouter sessions are added)
+- `$board` history
+- Engineering-ready handoff (story IDs, epics, dependencies)
+
+## Example — TeamFlow AI 1:1 Summary: Q2 Export feature (Spec)
+
+**Request:** "Scope approved for Q2: add PDF/Notion export for summaries. A PRD is needed for the engineering team."
+
+**Mode:** /spec (strategy + scope already approved; specification-only work)
+
+**Session 1:**
+
+- **COND-01**: Scope validation — Q2 roadmap theme "Export", scope includes PDF + Notion, out-of-scope Slack. Timeline 3 weeks. WCAG AA. 5 questions answered.
+
+- **PM-SPEC**:
+  - PRD 12 pages: Problem (14 support tickets in Q3 asking for export), audience (end-user Manager), solution (native PDF + Notion API).
+  - 5 user stories INVEST: "As a manager I want to export summary as PDF so I can share with report". AC Gherkin per story.
+  - NFR: performance p95 < 3s for PDF gen; Notion API rate limit handling.
+  - Risks: Notion API deprecation risk (mitigation: vendored client); PDF rendering quality (mitigation: preview in UI).
+  - Rollout: opt-in beta for 10 managers, GA at 4.0 rating.
+  - Dependencies: Notion integration (new), PDF service (existing).
+
+- **UX-01**:
+  - User flow: Click "Export" on summary → select format (PDF/Notion) → download/open Notion page.
+  - Low-fi wireframes: 2 screens (summary detail + export modal).
+  - Accessibility: keyboard nav for modal, aria-label for format selector.
+
+- **TECH-01**:
+  - Feasibility: Yes. T-shirt = M (2 team-weeks).
+  - NFR: PDF gen via existing service, Notion integration new (1 week spike for auth OAuth flow).
+  - Risks: Notion API rate limit (mitigation: batch + retry), PDF large file size (mitigation: compression).
+  - Epics: 2 (PDF pipeline, Notion integration).
+
+- **DATA-01 (N/A)**: Metrics inherited — "export events" logged via existing event pipeline. Skip.
+
+- **LITE-RG**:
+  - PRD checklist: 11/12 ✅ (open questions listed, 1 noted "Notion Teams API vs Personal").
+  - User sign-off received.
+
+**Result:** 1 session × 3h = ready PRD 12 pages, engineering-ready.
+
+---
+
+## Anti-patterns
+
+| Mistake | Why it's bad | How to do it right |
+|---------|-------------|--------------------|
+| PRD without success metrics | Impossible to assess win | NSM / OKR link is mandatory |
+| Implementation details in PRD | PRD = what/why, not how | How → engineering docs |
+| User stories without AC | AC — contract with engineers | Gherkin Given/When/Then |
+| Skipping NFR | In B2B SaaS often deal-breakers | NFR top 5 minimum |
+| Scope reveal in spec session | Violates pipeline order | Escalate — return to /shape-prioritize or /ship-right-thing |
+| No rollout plan | Risk of incidents at GA | Pilot → Beta → GA with criteria + rollback |
+| Open questions without owner | Blocked during impl | Owner + deadline per question |

package/domains/product/locales/en/AGENTS.md

@@ -0,0 +1,149 @@
+# Product Domain Orchestrator
+
+## Overview
+The product domain is an agent system for **B2B SaaS product management**: from discovery to release. Two adversarial axes provide decision stability:
+- **Customer-Champion vs Business-Champion** — at the strategic level (why and for whom).
+- **Build-Camp vs Cut-Camp** — at prioritization (how much and what scope).
+
+Mediator synthesizes the results of both camps. Designer + Layouter prepare the PRD / Product Review Deck as PDF.
+
+## Pipeline A — `/ship-right-thing` (Customer-vs-Business, 6 sessions)
+```
+Session 1: CONDUCTOR → DISCOVERY                                        → Handoff
+Session 2: CONDUCTOR → CUSTOMER-CHAMPION (product_strategist α)         → Handoff
+Session 3: CONDUCTOR → BUSINESS-CHAMPION (product_strategist β)         → Handoff
+Session 4: CONDUCTOR → MEDIATOR                                         → Handoff
+Session 5: CONDUCTOR → PM → UX_DESIGNER + TECH_LEAD + DATA_ANALYST      → Handoff
+Session 6: CONDUCTOR → DESIGNER → LAYOUTER                              → PDF
+```
+
+## Pipeline B — `/shape-prioritize` (Build-vs-Cut, 6 sessions)
+Triggered when strategy is already approved — discussion focuses on scope.
+```
+Session 1: CONDUCTOR → DISCOVERY (scope intake)                         → Handoff
+Session 2: CONDUCTOR → BUILD-CAMP (pm α + tech_lead α)                  → Handoff
+Session 3: CONDUCTOR → CUT-CAMP (pm β + tech_lead β)                    → Handoff
+Session 4: CONDUCTOR → MEDIATOR                                         → Handoff
+Session 5: CONDUCTOR → PM → UX_DESIGNER + DATA_ANALYST                  → Handoff
+Session 6: CONDUCTOR → DESIGNER → LAYOUTER                              → PDF (PRD)
+```
+
+## Pipeline Quick — `/quick-pm` (1 session, no adversarial)
+```
+CONDUCTOR → DISCOVERY → product_strategist → PM → TECH_LEAD → DATA_ANALYST → PDF
+```
+
+## Pipeline Spec — `/spec` (1 session, PRD only)
+```
+CONDUCTOR → PM → UX_DESIGNER + TECH_LEAD → PRD (markdown)
+```
+
+## Teams (adversarial)
+
+| Team | Active in | Composition | Task |
+|---------|-----------|--------|--------|
+| Customer-Champion (Alpha) | /ship-right-thing | product_strategist α | Strategy around user value, JTBD, adoption |
+| Business-Champion (Beta) | /ship-right-thing | product_strategist β | Strategy around revenue, margin, strategic moats |
+| Build-Camp (Alpha) | /shape-prioritize | pm α + tech_lead α | Maximum scope: every feature is justified |
+| Cut-Camp (Beta) | /shape-prioritize | pm β + tech_lead β | Minimum scope: cut everything that can be cut |
+| Layout Team | all pipelines with PDF | Designer → Layouter | Layout and typesetting of PRD / Product Review Deck |
+
+## Agents
+
+| Role | File | Area of Responsibility |
+|------|-------|---------------------|
+| Conductor | agents/conductor.md | Orchestration, session routing, gate control |
+| Discovery | agents/discovery.md | Scope, user research intake, jobs, assumptions |
+| Product Strategist | agents/product_strategist.md | Vision, OKR, NSM, roadmap. Plays both sides in Customer-vs-Business |
+| PM | agents/pm.md | Prioritization, PRD, stories. Plays both sides in Build-vs-Cut |
+| UX Designer | agents/ux_designer.md | Wireframes, user flows, design brief |
+| Tech Lead | agents/tech_lead.md | Feasibility, breakdown, risk, non-functional requirements |
+| Data Analyst | agents/data_analyst.md | Experiment design, SaaS metrics, AARRR, hypothesis validation |
+| Mediator | agents/mediator.md | Impartial synthesis of Customer/Business and Build/Cut |
+| Designer | agents/designer.md | PRD / Product Review Deck layout, visualizations |
+| Layouter | agents/layouter.md | HTML/CSS → PDF, charts, document assembly |
+
+## Workflow
+
+| Command | Mode | Adversarial Axis |
+|---------|-------|-----------------|
+| /ship-right-thing | Full Pipeline A (6 sessions) | Customer-Champion vs Business-Champion |
+| /shape-prioritize | Full Pipeline B (6 sessions) | Build-Camp vs Cut-Camp |
+| /spec | PRD Only (1 session) | — |
+| /quick-pm | Quick Pipeline (1 session) | — |
+
+## Skills by Agent
+
+### Conductor
+- $board — task board
+- $handoff — handoff between gates/sessions
+- $gates — gate control
+- $session-prompt-generator — next session prompt
+
+### Discovery
+- $user-interview-script — B2B interview script
+- $jtbd-canvas — Jobs-to-be-Done
+- $opportunity-solution-tree — opportunity-solution tree
+- $problem-statement — problem statement
+- $assumption-mapping — assumption map
+- $board, $handoff
+
+### Product Strategist
+- $product-vision — product vision
+- $okr-framework — OKR
+- $north-star-metric — NSM
+- $product-roadmap — product roadmap
+- $kano-model — Kano model
+- $handoff
+
+### PM
+- $rice-scoring — RICE
+- $moscow-prioritization — MoSCoW
+- $wsjf-scoring — WSJF (for enterprise)
+- $prd-template — PRD
+- $user-story — user stories
+- $acceptance-criteria — acceptance criteria
+- $epic-breakdown — epic breakdown
+- $kano-model — Kano model (for scope discussions)
+- $handoff
+
+### UX Designer
+- $design-brief — design brief
+- $user-flow — user flows
+- $handoff
+
+### Tech Lead
+- $epic-breakdown — epic breakdown
+- $handoff
+
+### Data Analyst
+- $hypothesis-template — hypothesis
+- $ab-test-design — A/B test design
+- $saas-metrics — B2B SaaS metrics
+- $aarrr-metrics — AARRR
+- $north-star-metric — NSM (for metric design)
+- $handoff
+
+### Mediator
+- $handoff — receiving results from both camps
+- $gates — final check
+
+### Designer
+- $report-design — PRD / Product Review Deck layout
+
+### Layouter
+- $html-pdf-report — HTML/CSS → PDF
+
+### Release (cross-agent skills)
+- $launch-checklist — launch checklist
+- $gtm-brief — GTM brief (handoff to marketing/sales)
+- $release-notes — release notes for B2B clients
+
+## MCP Integrations (expected)
+- **mempalace** — long-term product memory, decisions, discovery artifacts
+- **shadcn + tailwindcss + stitch** — for UI mockups via UX designer
+- **scheduled-tasks** — discovery cadence, regular reviews
+
+## Localization
+- Default: `ru`
+- Available: `ru` (v1.0), `en` (v1.1+)

package/domains/product/locales/en/AGENTS.yaml

@@ -0,0 +1,78 @@
+version: 1
+name: "product_management_orchestra"
+display_name: "AGENTS.md — Product Management Orchestra"
+description: "Orchestration contract for roles, skills, and delivery gates for B2B SaaS product management with dual adversarial axes (Customer-vs-Business, Build-vs-Cut)."
+default_prompt: "Use AGENTS.md when the task requires orchestrating the product pipeline."
+triggers:
+  - "AGENTS.md"
+  - "product orchestrator"
+  - "product pipeline"
+  - "product management"
+  - "B2B SaaS"
+  - "PRD"
+  - "roadmap"
+  - "discovery"
+roles:
+  - "conductor"
+  - "discovery"
+  - "product_strategist"
+  - "pm"
+  - "ux_designer"
+  - "tech_lead"
+  - "data_analyst"
+  - "mediator"
+  - "designer"
+  - "layouter"
+skills:
+  - "ab-test-design"
+  - "aarrr-metrics"
+  - "acceptance-criteria"
+  - "assumption-mapping"
+  - "board"
+  - "design-brief"
+  - "epic-breakdown"
+  - "gates"
+  - "gtm-brief"
+  - "handoff"
+  - "html-pdf-report"
+  - "hypothesis-template"
+  - "jtbd-canvas"
+  - "kano-model"
+  - "launch-checklist"
+  - "moscow-prioritization"
+  - "north-star-metric"
+  - "okr-framework"
+  - "opportunity-solution-tree"
+  - "prd-template"
+  - "problem-statement"
+  - "product-roadmap"
+  - "product-vision"
+  - "release-notes"
+  - "report-design"
+  - "rice-scoring"
+  - "saas-metrics"
+  - "session-prompt-generator"
+  - "user-flow"
+  - "user-interview-script"
+  - "user-story"
+  - "wsjf-scoring"
+pipeline:
+  - "CONDUCTOR"
+  - "DISCOVERY"
+  - "PRODUCT_STRATEGIST"
+  - "PM"
+  - "UX_DESIGNER"
+  - "TECH_LEAD"
+  - "DATA_ANALYST"
+  - "MEDIATOR"
+  - "DESIGNER"
+  - "LAYOUTER"
+  - "RELEASE_GATE"
+invocation:
+  explicit: false
+  implicit: true
+localization:
+  default_locale: "en"
+  available_locales:
+    - "ru"
+    - "en"

package/domains/product/locales/en/CONTEXT.md

@@ -0,0 +1,40 @@
+# Project Overview
+
+| Field | Value |
+|------|----------|
+| **Domain** | Product Management (B2B SaaS) |
+| **Description** | Agent system for product management with two adversarial axes |
+| **Current Phase** | MVP |
+
+## Pipeline A Results — `/ship-right-thing` (Customer-vs-Business)
+
+| Session | Gate | Status | Key Result | Date |
+|--------|------|--------|---------------------|------|
+| 1 | Conductor | ☐ | — | — |
+| 1 | Discovery | ☐ | — | — |
+| 2 | Customer-Champion: Strategist | ☐ | — | — |
+| 3 | Business-Champion: Strategist | ☐ | — | — |
+| 4 | Mediator (Strategy Synthesis) | ☐ | — | — |
+| 5 | PM (prioritization + PRD) | ☐ | — | — |
+| 5 | UX Designer (wireframes) | ☐ | — | — |
+| 5 | Tech Lead (feasibility) | ☐ | — | — |
+| 5 | Data Analyst (metric plan) | ☐ | — | — |
+| 6 | Designer | ☐ | — | — |
+| 6 | Layouter | ☐ | — | — |
+| 6 | Release Gate | ☐ | — | — |
+
+## Pipeline B Results — `/shape-prioritize` (Build-vs-Cut)
+
+| Session | Gate | Status | Key Result | Date |
+|--------|------|--------|---------------------|------|
+| 1 | Conductor | ☐ | — | — |
+| 1 | Discovery (scope intake) | ☐ | — | — |
+| 2 | Build-Camp: PM + Tech Lead | ☐ | — | — |
+| 3 | Cut-Camp: PM + Tech Lead | ☐ | — | — |
+| 4 | Mediator (Scope Synthesis) | ☐ | — | — |
+| 5 | PM (final PRD) | ☐ | — | — |
+| 5 | UX Designer | ☐ | — | — |
+| 5 | Data Analyst | ☐ | — | — |
+| 6 | Designer | ☐ | — | — |
+| 6 | Layouter | ☐ | — | — |
+| 6 | Release Gate | ☐ | — | — |