code-ai-installer 2.2.0 → 2.3.0

package/domains/product/locales/en/.agents/skills/product-vision/SKILL.md

@@ -0,0 +1,395 @@
+---
+name: product-vision
+description: Product vision — for whom / why / what's unique / 2-3 year horizon / guiding principles
+---
+# Product Vision
+
+> **Category:** Strategy  ·  **Slug:** `product-vision`
+
+## When to Use
+
+- At the start of a new product direction (new product line, pivot).
+- When onboarding new teams — vision as a shared north star.
+- When the team diverges on strategic decisions — no shared framework.
+- For investor / exec communication — vision = story.
+
+## Input
+
+| Field | Required | Description |
+|-------|:--------:|-------------|
+| Discovery Brief (JTBD, problems) | ✅ | Customer-value foundation |
+| Business context (ARR, segment, stage) | ✅ | Business-value foundation |
+| Competitive landscape | ⬚ | Differentiation context |
+| Horizon | ⬚ | 2-3 years is standard |
+| Adversarial camp (Full A) | ⬚ | alpha (Customer) / beta (Business) |
+
+## Data Sources
+
+1. `$jtbd-canvas` — language for whom / what job.
+2. Business metrics (ARR, NRR, churn) — for the Business-Champion lens.
+3. Competitive analysis — moat identification.
+4. Company mission / values — alignment.
+
+### Related Skills
+
+| Skill | What we take | When to invoke |
+|-------|-------------|----------------|
+| `jtbd-canvas` | Customer frame | For user-value formulation |
+| `okr-framework` | Vision → quarterly goals | After vision |
+| `north-star-metric` | Vision → measurable NSM | After vision |
+| `product-roadmap` | Vision → themes | After NSM |
+
+## Vision Formula (Geoffrey Moore)
+
+```
+For [target customer]
+Who [statement of the need or opportunity]
+Our [product name] is a [product category]
+That [statement of key benefit — that is, compelling reason to buy]
+Unlike [primary competitive alternative]
+Our product [statement of primary differentiation]
+```
+
+## Protocol
+
+### Step 0 — Role Determination (Full A)
+
+If Full Pipeline A — determine the camp (Customer-Champion or Business-Champion). Different emphases:
+
+**Customer-Champion:**
+- For whom — specific end-user / buyer segment
+- Why — user outcome, JTBD fulfillment
+- Differentiator — user experience, value per usage
+
+**Business-Champion:**
+- For whom — segment with highest NRR / expansion potential
+- Why — business outcome (ARR, moat, strategic asset)
+- Differentiator — technical/market moat, scale economics
+
+### Step 1 — Target Customer
+
+Specific segment, not «B2B companies»:
+- Firmographic (size, industry, geography)
+- Psychographic (buying behavior, maturity)
+- Primary role (buyer) + primary user (end-user)
+
+### Step 2 — Need / Opportunity
+
+One **primary** unmet job or problem (from Discovery). Not a list — ONE.
+
+### Step 3 — Product Category
+
+Which category we play in. If a new category — document what it includes.
+
+### Step 4 — Key Benefit
+
+Outcome-focused («get X done in Y time») or emotional («confident about Z»), not a feature list.
+
+### Step 5 — Alternative
+
+Honestly: what do customers currently use instead of our product? Including «do nothing» and self-built solutions.
+
+### Step 6 — Differentiation
+
+Why are we better for this customer in this job? Not «better at X» generally — a specific differentiator.
+
+### Step 7 — Guiding Principles (3-5)
+
+Points that guide decisions:
+- «Ship small and often over perfect and late»
+- «Optimize for end-user time-to-value, not feature count»
+- «Enterprise-grade, SMB-simple»
+
+### Step 8 — Vision Story
+
+Final artifact: 1-3 paragraphs, reads as a story. Includes:
+- Who's the hero (target customer)
+- What's their pain
+- How our product changes their life
+- Where we'll be in 2-3 years
+
+### Step 9 — Vision Test (FOCUS)
+
+- **F**easible — can we implement it?
+- **O**bsession-worthy — does it inspire the team?
+- **C**larifying — does it help make trade-offs?
+- **U**nique — different from competitors' copies?
+- **S**pecific — not generic words?
+
+## Validation (Quality Gate)
+
+- [ ] Geoffrey Moore formula filled in
+- [ ] Target customer — specific, not generic
+- [ ] Need — one primary job, not a list
+- [ ] Differentiation — specific, checkable
+- [ ] 3-5 guiding principles
+- [ ] Vision story ≤ 3 paragraphs
+- [ ] Passes FOCUS test
+- [ ] (Full A) — camp determined, emphases match
+
+## Handoff
+
+The result is input for:
+- **`okr-framework`** — vision → quarterly Objectives
+- **`north-star-metric`** — vision → measurable NSM
+- **`product-roadmap`** — vision → themes
+- **Mediator** (Full A) — alpha and beta vision for synthesis
+
+Format: vision doc (~1 page) + Moore-formula + principles. Via `$handoff`.
+
+## Anti-patterns
+
+| Error | Why it's bad | How to do it right |
+|-------|-------------|-------------------|
+| Vision = feature roadmap | No differentiator, no story | Story: hero + pain + transform |
+| Generic customer | «Teams» | Specific: «DevOps teams in mid-market SaaS» |
+| Multiple «primary» needs | Distraction | ONE primary, rest secondary |
+| Copy competitors | No moat | Specific differentiator |
+| Feasible only in 10 years | Team disengages | 2-3 years horizon |
+| No guiding principles | Team trade-offs by mood | 3-5 principles |
+
+## Template
+
+```markdown
+# Product Vision
+
+## Target Customer
+[Specific segment]
+
+## Need / Opportunity
+[ONE primary unmet job / problem]
+
+## Product Category
+[Category we play in]
+
+## Key Benefit
+[Compelling outcome]
+
+## Unlike (Alternative)
+[Current alternatives + why we're different]
+
+## Differentiation
+[Specific, defensible]
+
+## Guiding Principles
+1. ...
+
+## Vision Story (1-3 paragraphs)
+...
+
+## FOCUS Check
+- Feasible: ✅
+- Obsession-worthy: ✅
+- Clarifying: ✅
+- Unique: ✅
+- Specific: ✅
+```
+
+## Worked Example — TeamFlow (Full Pipeline A: Customer-Champion vs Business-Champion)
+
+**Context:** TeamFlow AI summarization initiative. Full Pipeline A (`/ship-right-thing`) — adversarial vision sessions. Alpha (STRAT-01α) and Beta (STRAT-02β) work **independently**, each receiving only the Discovery Brief. The Mediator synthesizes in session 4. Below — complete vision docs from both camps + Mediator output.
+
+### Camp Alpha — Customer-Champion Vision (STRAT-01α)
+
+```markdown
+# Product Vision — AI 1:1 Summarization (Customer-Champion perspective)
+
+## Target Customer
+People managers (5-15 direct reports) in B2B SaaS / Tech companies 100-1000 employees.
+Secondary: individual contributors, benefiting indirectly through better managers.
+
+## Need / Opportunity
+Every week managers lose 3-4 hours to admin overhead (prep + note-taking + follow-up).
+This robs both time AND attention from actual conversation — when a manager types notes during
+a 1:1, they don't listen deeply. End-user managers want TeamFlow to **handle the admin work
+so they can focus on people**.
+
+## Product Category
+AI-assisted 1:1 workflow (new category — extends «performance management platform»).
+
+## Key Benefit
+**Reclaim 3 hours per week of deep work time + transform 1:1s from anxiety-inducing
+admin task into confident, connected conversations.** Manager enters each 1:1
+knowing context, leaves without homework, never loses an action item.
+
+## Unlike (Alternatives)
+- **Manual workflow (today):** 3-4 hrs/week lost, 40% action items dropped, anxiety
+  + guilt from skipped prep.
+- **Otter.ai + ChatGPT workaround:** privacy concern (confidential HR data in OpenAI
+  consumer), disconnected from TeamFlow context.
+- **Lattice / 15Five:** same gap — no native AI summarization. Manual templates only.
+
+## Differentiation
+**Privacy-first AI integrated in TeamFlow's 1:1 data model.** Summaries know previous
+1:1s, action items, OKR context. Standalone transcription tools can't replicate
+contextual awareness. Competitors may catch up on AI — but our end-user trust moat
+(3+ years of HR data handling) is deeper.
+
+## Guiding Principles
+1. **End-user time is sacred.** Every feature must save minutes, not add them.
+2. **Summaries supplement, not replace, human judgment.** Inline edit always available.
+3. **Privacy by design.** Manager controls every toggle. No training on user data.
+4. **Integrated always beats standalone.** Context > accuracy.
+5. **Delight the manager, not surveil them.** Individual value, not management control.
+
+## Vision Story
+Imagine Marcus, an engineering manager with 8 reports. Monday morning: TeamFlow shows
+him an AI-generated prep card for his 10am 1:1 with Sarah — last week's commitments
+resurfaced, her recent Slack mood indicators flagged, suggested discussion prompts
+from org patterns. During the conversation, Marcus is fully present — no typing.
+When it ends, TeamFlow has a draft summary ready, action items extracted, ready for
+Marcus's one-tap approval. By Friday, Marcus has spent 45 minutes on 1:1 admin
+instead of 3.5 hours. He's a better manager because the tool does the work AI does best.
+
+By 2028, TeamFlow has become the standard: if you're a manager and you're not using
+AI-assisted 1:1s, you're considered behind — the way typing without autocomplete feels now.
+
+## FOCUS Check
+- **Feasible:** ✅ LLM APIs at required quality exist today
+- **Obsession-worthy:** ✅ Managers personally want this, beyond business case
+- **Clarifying:** ✅ "Does this save manager time?" — clear decision filter
+- **Unique:** ✅ Native TeamFlow integration + privacy architecture
+- **Specific:** ✅ Concrete segment, concrete time-save, concrete alternatives
+```
+
+---
+
+### Camp Beta — Business-Champion Vision (STRAT-02β)
+
+```markdown
+# Product Vision — AI 1:1 Summarization (Business-Champion perspective)
+
+## Target Customer
+VP People / Chief People Officers at mid-market and enterprise B2B SaaS companies
+($5M-$500M ARR, 100-1000 employees). Secondary: managers, as tool users.
+
+## Need / Opportunity
+Mid-market CPOs cannot **measure or drive** manager effectiveness at scale. They rely
+on storytelling + anecdata when CFO asks «what's the ROI of $200K manager training?».
+The gap: no data layer on 1:1 practice. **We unlock measurable People Ops** — turning
+HR into a data-driven discipline on par with Sales and Engineering.
+
+## Product Category
+People Ops analytics platform (category expansion from «performance management
+platform» to broader «people operations intelligence»).
+
+## Key Benefit
+**Make People Operations as measurable as Sales Operations.** Dashboard showing 1:1
+cadence, action items velocity, team health indicators per-manager and org-wide —
+the metrics CPOs have always wanted. Turns the «soft» HR function into a credible
+ELT presence.
+
+## Unlike (Alternatives)
+- **Culture Amp, Lattice surveys:** lagging, quarterly, employee-reported. Our metrics
+  are leading, continuous, behavioral.
+- **Spreadsheets + tribal knowledge:** doesn't scale past 200 employees.
+- **Zendesk-style approach «just count tickets»:** doesn't surface cause.
+
+## Differentiation
+**Leading indicators of manager effectiveness**, extracted from 1:1 patterns. Moat: 
+1. 3+ years of customer 1:1 behavioral data (training corpus for AI)
+2. Network effect — more customers → better aggregate benchmarks (is your 1:1 cadence
+   in top quartile?)
+3. Analytics layer as expansion driver (Enterprise tier unlocks this)
+
+## Guiding Principles
+1. **Measure or it doesn't exist.** Every feature generates a data point.
+2. **Analytics drives tier expansion.** Basic usage = Team tier, analytics = Enterprise.
+3. **Aggregate > individual for buyers.** Never surveillance content.
+4. **Benchmark as value-add.** Industry / cohort comparisons = hard-to-copy.
+5. **ROI visibility for CPO.** Dashboard → Board deck in 1 click.
+
+## Vision Story
+Imagine Jennifer, CPO at a 450-person B2B SaaS. Monday morning, she reviews the TeamFlow
+dashboard: 89% of managers ran 1:1s last week (vs 82% last month — green arrow). Action
+item completion rate is 68% org-wide, but the Engineering team is at 52% (flag: manager
+training opportunity). Her top-performer-retention cohort shows a 1:1 pattern that differs
+from the at-risk cohort — 3 weeks of early warning before regrettable attrition.
+
+When CEO asks «what's ROI on People investment?», Jennifer has a one-page dashboard
+ready: 1:1 cadence correlates with retention r=0.71, actions velocity correlates
+with NPS r=0.58. For the first time, People shows up at ELT with hard numbers.
+
+By 2028, TeamFlow is the «Salesforce for People Operations» — the platform where
+CPOs spend their morning like CROs live in their pipeline dashboard. Category-defining,
+not feature-expanding.
+
+## FOCUS Check
+- **Feasible:** ✅ Data exists in product, surfacing is eng-solvable
+- **Obsession-worthy:** ✅ CPOs have been asking for this for years
+- **Clarifying:** ✅ "Does this give CPOs a new metric?" — filter
+- **Unique:** ✅ Our data corpus is unique; benchmarks create moat
+- **Specific:** ✅ Concrete persona, concrete ELT narrative, concrete metrics
+```
+
+---
+
+### Mediator Synthesis (MED-01, Session 4)
+
+**Alpha evidence strength:** 4/5 (strong JTBD connection, strong competitive differentiation)  
+**Beta evidence strength:** 4/5 (strong buyer need, strong tier expansion rationale)  
+
+**Key disagreements:**
+1. **Factual:** Alpha says «manager time saved drives retention»; Beta says «analytics drives expansion». Evidence supports BOTH — not mutually exclusive.
+2. **Value:** Alpha centers end-user; Beta centers buyer. Both are real customers in the B2B buying loop.
+3. **Risk:** Alpha under-weights aggregate dashboard importance for Enterprise; Beta under-weights privacy concerns for SMB.
+
+**Synthesis: HYBRID vision**
+
+```markdown
+# Unified Product Vision — AI-Assisted 1:1 Workflow + Analytics
+
+## Target Customer (Expanded)
+Two user types, both buyers in B2B People Ops decisions:
+- **Primary user:** People managers (5-15 reports)
+- **Primary buyer:** VP People / CPO (mid-market + enterprise)
+
+## Need / Opportunity
+Simultaneously solve two connected problems:
+1. Managers lose 3-4 hrs/week to 1:1 admin (end-user pain)
+2. CPOs cannot measure manager effectiveness (buyer pain)
+Both stem from the same root: 1:1 workflow is manual, leaving no data and no time.
+
+## Key Benefit
+**For managers:** Reclaim hours per week, transform 1:1s from admin to connection.
+**For CPOs:** Make People Ops measurable — leading indicators + benchmarks.
+Both benefits reinforce each other: time saved → more 1:1s happen → more data →
+better insights → better coaching → retention.
+
+## Product Category
+AI-powered People Operations platform (convergence of workflow + analytics tiers).
+
+## Differentiation (Moat Stack)
+1. **Privacy architecture** (from Alpha): end-user trust
+2. **Behavioral data corpus** (from Beta): analytics moat
+3. **Integrated workflow** (from Alpha): context-aware summaries
+4. **Benchmark network** (from Beta): industry comparison value
+
+## Guiding Principles (Merged)
+1. End-user time is sacred AND generates valuable data — alignment.
+2. Summaries are individual; analytics are aggregate. Never mix.
+3. Privacy by design (end-user consent for AI; aggregate-only for buyer).
+4. Tiered value: Team tier = workflow, Enterprise tier = analytics.
+5. Delight manager AND equip CPO — both stakeholders win.
+
+## Tier Architecture (from synthesis)
+- **Core (today):** Manual 1:1 workflow
+- **AI Team tier (+$8/seat):** AI summarization, auto action items, reminders (Alpha focus)
+- **Enterprise tier (+$15/seat):** Aggregate dashboard, manager health scores, benchmarks (Beta focus)
+
+## Open Risks
+- [Alpha concern] Privacy architecture must be bulletproof for end-user adoption
+- [Beta concern] Analytics tier must not feel like surveillance (Alpha principle #5 applied)
+- [Mediator add] Pricing can cannibalize if not tiered cleanly — test with customer advisory board
+
+## Rationale for Hybrid
+Alpha and Beta are not competing visions — they're complementary layers of the same
+platform. Choosing only one would:
+- Alpha only: miss $2-3M ARR Enterprise expansion opportunity
+- Beta only: lack end-user love, leading to low adoption and churn risk
+
+Synthesis: **Ship Alpha MVP first (workflow), layer Beta (analytics) as Enterprise
+tier within 4-6 months.** Sequencing respects end-user trust (must work before surveilling).
+```
+
+> **Product-vision adversarial lesson:** Customer vs Business camps, independent work, Mediator synthesis. **Neither camp won outright** — the synthesis revealed both were right but incomplete. Without the formal adversarial process, the team would have shipped Alpha's vision alone and missed the $2-3M Enterprise ARR opportunity, OR shipped Beta's vision alone and faced adoption backlash. The structured debate cost 2 extra sessions but saved at least one strategic misstep.

package/domains/product/locales/en/.agents/skills/product-vision/agents/claude.json

@@ -0,0 +1,15 @@
+{
+  "name": "product-vision",
+  "display_name": "Product Vision",
+  "description": "Product vision — for whom / why / what's unique / 2-3 year horizon / guiding principles",
+  "default_prompt": "Use $product-vision when the task relates to the skill \"Product Vision\".",
+  "triggers": [
+    "product-vision",
+    "product vision"
+  ],
+  "capabilities": [
+    "product-vision"
+  ],
+  "tools": [],
+  "implicit_invocation": true
+}

package/domains/product/locales/en/.agents/skills/product-vision/agents/copilot.json

@@ -0,0 +1,15 @@
+{
+  "name": "product-vision",
+  "display_name": "Product Vision",
+  "description": "Product vision — for whom / why / what's unique / 2-3 year horizon / guiding principles",
+  "default_prompt": "Use $product-vision when the task relates to the skill \"Product Vision\".",
+  "triggers": [
+    "product-vision",
+    "product vision"
+  ],
+  "capabilities": [
+    "product-vision"
+  ],
+  "tools": [],
+  "implicit_invocation": true
+}

package/domains/product/locales/en/.agents/skills/product-vision/agents/gemini.json

@@ -0,0 +1,15 @@
+{
+  "name": "product-vision",
+  "display_name": "Product Vision",
+  "description": "Product vision — for whom / why / what's unique / 2-3 year horizon / guiding principles",
+  "default_prompt": "Use $product-vision when the task relates to the skill \"Product Vision\".",
+  "triggers": [
+    "product-vision",
+    "product vision"
+  ],
+  "capabilities": [
+    "product-vision"
+  ],
+  "tools": [],
+  "implicit_invocation": true
+}

package/domains/product/locales/en/.agents/skills/product-vision/agents/openai.yaml

@@ -0,0 +1,8 @@
+interface:
+  display_name: "Product Vision"
+  short_description: "Product vision — for whom / why / what's unique / 2-3 year horizon / guiding principles"
+  default_prompt: "Use $product-vision when the task relates to the skill \"Product Vision\"."
+dependencies:
+  tools: []
+policy:
+  allow_implicit_invocation: true

package/domains/product/locales/en/.agents/skills/product-vision/agents/qwen.json

@@ -0,0 +1,15 @@
+{
+  "name": "product-vision",
+  "display_name": "Product Vision",
+  "description": "Product vision — for whom / why / what's unique / 2-3 year horizon / guiding principles",
+  "default_prompt": "Use $product-vision when the task relates to the skill \"Product Vision\".",
+  "triggers": [
+    "product-vision",
+    "product vision"
+  ],
+  "capabilities": [
+    "product-vision"
+  ],
+  "tools": [],
+  "implicit_invocation": true
+}

package/domains/product/locales/en/.agents/skills/product-vision/agents/skill.yaml

@@ -0,0 +1,18 @@
+version: 1
+name: "product-vision"
+display_name: "Product Vision"
+description: "Product vision — for whom / why / what's unique / 2-3 year horizon / guiding principles"
+default_prompt: "Use $product-vision when the task relates to the skill \"Product Vision\"."
+triggers:
+  - "product-vision"
+  - "product vision"
+capabilities:
+  - "product-vision"
+tools: []
+invocation:
+  explicit: true
+  implicit: true
+localization:
+  default_locale: "en"
+  available_locales:
+    - "en"