code-ai-installer 2.1.0 → 2.3.0

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

@@ -0,0 +1,419 @@
+---
+name: launch-checklist
+description: Launch checklist — internal (dogfooding, training, rollback) + external (GTM, customer comms)
+---
+# Launch Checklist (B2B SaaS)
+
+> **Category:** Release  ·  **Slug:** `launch-checklist`
+
+## When to Use
+
+- 2-4 weeks before launching a feature / product / tier.
+- For phased rollout planning.
+- For alignment Product + Eng + CS + Sales + Marketing.
+- As a quality gate before GA.
+
+## Input
+
+| Field | Required | Description |
+|-------|:--------:|-------------|
+| Feature / initiative | ✅ | What we're launching |
+| Target launch date | ✅ | Planned date |
+| Stakeholders | ✅ | PM, Eng, CS, Sales, Marketing, Legal |
+| Customer impact level | ✅ | Minor / Major / Breaking |
+| Regulatory implications | ⬚ | GDPR, industry-specific |
+
+## Data Sources
+
+1. `$prd-template` — feature scope.
+2. `$gtm-brief` — external messaging.
+3. `$release-notes` — customer-facing announcement.
+4. `$rollback-plan` (if separate) — contingency.
+
+### Related Skills
+
+| Skill | What we take | When to invoke |
+|-------|-------------|----------------|
+| `prd-template` | What we're launching | Parent input |
+| `gtm-brief` | External messaging | Part of launch |
+| `release-notes` | Customer-facing | Part of launch |
+| `ab-test-design` | Phased rollout design | If A/B launch |
+
+## Launch Phases
+
+1. **T-4 weeks: Internal prep**
+2. **T-2 weeks: Soft launch (dogfood + beta customers)**
+3. **T-1 week: Pre-launch hardening**
+4. **T-0: Launch day**
+5. **T+1 week: Monitor + iterate**
+6. **T+4 weeks: Post-launch review**
+
+## Checklist
+
+### 🔧 Internal Preparation (T-4 weeks)
+
+**Engineering:**
+- [ ] Feature flag implemented + tested
+- [ ] Observability: metrics dashboard created
+- [ ] Alerts configured for error rate, latency, usage
+- [ ] Load testing done (for high-traffic features)
+- [ ] Security review completed
+- [ ] Rollback tested (revert works end-to-end)
+
+**Product:**
+- [ ] PRD finalized
+- [ ] Success metrics defined + instrumented
+- [ ] Internal demo ready
+- [ ] Edge cases documented
+
+**Legal / Compliance:**
+- [ ] ToS / Privacy policy updates (if needed)
+- [ ] DPA updates (if data handling changes)
+- [ ] Industry compliance (SOC2, HIPAA, GDPR) reviewed
+- [ ] Export control considerations (if applicable)
+
+**Data:**
+- [ ] Analytics instrumentation live
+- [ ] Dashboard created for monitoring post-launch
+- [ ] Baseline metrics captured (for before/after comparison)
+
+### 🧪 Soft Launch (T-2 weeks)
+
+**Dogfooding:**
+- [ ] Internal team using feature 5+ days
+- [ ] Feedback collected from internal users
+- [ ] Top 3 friction points addressed or documented
+
+**Beta Customers:**
+- [ ] Beta group identified (5-20 customers, varied segments)
+- [ ] Beta agreement signed (if changes ToS)
+- [ ] Feedback survey + interview sessions
+- [ ] Support team briefed on potential issues
+
+**Iteration:**
+- [ ] Beta feedback review session
+- [ ] Critical issues resolved
+- [ ] Feature gates adjusted
+
+### 📢 Pre-Launch Hardening (T-1 week)
+
+**Customer Communications:**
+- [ ] Release notes finalized (via `$release-notes`)
+- [ ] Email announcement drafted
+- [ ] In-app announcement banner designed
+- [ ] Blog post / changelog published
+- [ ] Documentation (public knowledge base) updated
+- [ ] Video walkthrough (if major feature)
+
+**Sales Enablement:**
+- [ ] Sales deck updated
+- [ ] Battle cards (vs competitors) updated
+- [ ] Demo script / environment ready
+- [ ] Pricing update (if applicable) approved
+
+**CS / Support Enablement:**
+- [ ] Training session conducted
+- [ ] Internal FAQ / runbook created
+- [ ] Support ticket categories updated
+- [ ] Escalation path for launch issues
+
+**Marketing:**
+- [ ] Landing page updated / new one ready
+- [ ] Social media posts scheduled
+- [ ] PR outreach (if major)
+- [ ] Partnership announcements (if applicable)
+
+### 🚀 Launch Day (T-0)
+
+**Morning of launch:**
+- [ ] Final health check (metrics dashboard green)
+- [ ] Support team on-call
+- [ ] PM monitoring metrics (first 2 hours intensive)
+- [ ] Feature flag enabled (% rollout per plan: 5% → 25% → 100%)
+
+**Communications:**
+- [ ] Email sent to customer list
+- [ ] In-app banner activated
+- [ ] Blog post published
+- [ ] Changelog update live
+- [ ] Social media posts published
+- [ ] Sales team notified
+
+**Monitoring:**
+- [ ] Error rate within baseline
+- [ ] Latency within SLA
+- [ ] Usage adoption tracked
+- [ ] Support ticket volume monitored
+
+### 📊 Post-Launch Monitoring (T+1 week)
+
+**Daily:**
+- [ ] Metrics dashboard review
+- [ ] Support ticket trend review
+- [ ] NPS feedback check
+
+**End of week 1:**
+- [ ] Success metric check (baseline → current)
+- [ ] Adoption rate
+- [ ] Issues discovered + logged
+- [ ] Customer feedback themes
+
+### 🔄 Post-Launch Review (T+4 weeks)
+
+- [ ] Success metrics vs targets
+- [ ] Adoption analysis by segment
+- [ ] Support ticket themes
+- [ ] Churn / expansion impact
+- [ ] Lessons learned document
+- [ ] Roadmap adjustments based on learnings
+
+## Rollback Plan (Always Required)
+
+- [ ] Trigger conditions defined:
+  - Error rate > X% for Y minutes
+  - Latency > SLA for Y minutes
+  - Critical customer escalation
+  - Security incident
+- [ ] Rollback mechanism: feature flag disable (preferred) or git revert
+- [ ] Rollback execution time: ≤ 5 minutes target
+- [ ] Communication plan: who notifies customers, via what channel
+- [ ] Post-rollback: root cause analysis, retry plan
+
+## Validation (Quality Gate)
+
+- [ ] All checklist sections completed
+- [ ] Feature flag + rollback tested
+- [ ] Observability instrumented
+- [ ] Beta feedback addressed
+- [ ] Communications ready (email, blog, docs, social)
+- [ ] Sales + CS enabled
+- [ ] Rollback plan with trigger conditions
+- [ ] Post-launch review scheduled
+
+## Handoff
+
+The output is the input for:
+- **Release Gate** (Conductor) — sign-off
+- **Engineering** → deployment
+- **CS / Sales / Marketing** → external execution
+- **Data Analyst** → post-launch monitoring
+
+Format: launch checklist markdown + tracking board. Via `$handoff`.
+
+## Anti-patterns
+
+| Error | Why it's bad | How to do it right |
+|-------|-------------|---------------------|
+| Big-bang launch | High risk, no safety net | Phased rollout (5% → 100%) |
+| No rollback plan | Outage extends | Always have Plan B |
+| No beta | Discover issues in production | Beta 2+ weeks |
+| No CS training | Tickets swamp support | Internal training T-1 week |
+| No observability | Blind post-launch | Metrics + alerts before launch |
+| Ignore legal | Compliance violations | Legal review for any data/contract change |
+| One person owns all | Single point of failure | Clear owner per section |
+
+## Template
+
+See sections above — use as checklist with tickable checkboxes.
+
+## Worked Example — TeamFlow AI Summarization Launch Checklist (GA)
+
+**Context:** AI Summarization MVP after beta phase. Full GA scheduled June 30, 2026. This checklist is the final production rollout.
+
+```markdown
+# Launch Checklist: AI Summarization GA (June 30, 2026)
+
+**PM Owner:** Alex K.  ·  **Launch Commander:** Jordan M. (Design Lead + PM backup)
+**Rollback On-call:** Priya S. (Eng Lead) + Sam P. (SRE)
+**Launch Date:** 2026-06-30 09:00 AM PT (Tuesday — mid-week, midday for coverage)
+
+---
+
+## 🔧 Internal Preparation (T-4 weeks, completed June 2)
+
+### Engineering
+- [x] Feature flag «ai_summarization_ga» configured (LaunchDarkly)
+- [x] Observability dashboards live: latency, error rate, generation volume, provider health
+- [x] Alerts configured:
+  - p95 latency >90s for >15 min → PagerDuty
+  - Error rate >2% for >15 min → PagerDuty
+  - LLM provider failover event → Slack alert
+  - Generation volume >200% baseline → Slack (investigate surge)
+- [x] Load testing: 500 concurrent summaries generated w/o degradation
+- [x] Security review completed and attested (SOC 2 scope extension)
+- [x] Rollback tested end-to-end (flag disable → feature off in <10s)
+- [x] LLM provider fallback tested (primary offline → secondary takes over, 12s recovery)
+
+### Product
+- [x] PRD finalized + all stories marked Done
+- [x] Success metrics instrumented and validated in staging (match schema)
+- [x] Internal demo ready (5-min walkthrough video)
+- [x] Edge cases documented in runbook
+
+### Legal / Compliance
+- [x] ToS update published and notified to all customers (30 days notice — May 30)
+- [x] DPA updates sent to Enterprise customers (200/200 signed)
+- [x] SOC 2 Type II extension for AI features — attestation complete
+- [x] GDPR compliance validated (EU data stays in Azure West Europe)
+- [x] Data Processing Record updated
+
+### Data
+- [x] Instrumentation live: all events from flow schema firing
+- [x] Dashboard published to leadership + stakeholders
+- [x] Baseline metrics captured (pre-AI): activation, retention, NPS, churn
+
+---
+
+## 🧪 Soft Launch (T-2 weeks, May 16 - June 2)
+
+### Dogfooding (T-3 to T-2, May 16-23)
+- [x] Internal TeamFlow team used feature 14 days (100% participation)
+- [x] Feedback collected: 47 internal users, 12 issues raised
+- [x] Top 5 issues resolved, 7 logged for Q3 iteration
+
+### Beta Customers (T-2, May 23 - June 2)
+- [x] 10 design partner customers selected (4 SMB, 4 mid-market, 2 enterprise)
+- [x] Beta agreement amended with AI clause (all 10 signed)
+- [x] Weekly feedback survey + 2 interview sessions conducted
+- [x] Support team briefed on potential issues (runbook + training)
+
+### Iteration
+- [x] Beta feedback review session held May 26 (2 hours, all-team)
+- [x] **Critical issues resolved:**
+  - Issue A: Summary «voice» too corporate, felt impersonal → prompt tuned, NPS improved +8
+  - Issue B: Action items sometimes missed when said «maybe» — threshold adjusted
+- [x] Feature gates adjusted based on beta telemetry
+
+---
+
+## 📢 Pre-Launch Hardening (T-1 week, June 23-29)
+
+### Customer Communications
+- [x] Release notes finalized (drafted via `$release-notes`, reviewed by Legal)
+- [x] Email announcement drafted + sent to Alex K for approval (3 variants: SMB, mid-market, enterprise)
+- [x] In-app announcement banner designed (copy approved by PM + Marketing)
+- [x] Blog post «Introducing AI-Assisted 1:1s» published in drafts, scheduled for June 30 9AM
+- [x] Documentation (public knowledge base):
+  - «Getting started with AI Summarization» (admin + manager guides)
+  - «Privacy & Security FAQ»
+  - «API changes related to AI features»
+- [x] Video walkthrough recorded (3 min, scheduled for release day)
+
+### Sales Enablement
+- [x] Sales deck updated (5 new slides added: «AI Summarization», «Tier Upgrade Path», «Objection Handling»)
+- [x] Battle cards updated vs Lattice, 15Five (AI feature comparison)
+- [x] Demo script + sandbox environment ready
+- [x] Pricing update approved and in CRM (CPQ updated for AI Tier)
+
+### CS / Support Enablement
+- [x] Training session conducted: 90-min live (recorded, 12 of 15 attended live, 3 viewed recording)
+- [x] Internal FAQ / runbook published (47 FAQ items documented)
+- [x] Support ticket categories updated («AI Summary: Quality», «AI Summary: Latency», «AI Summary: Privacy»)
+- [x] Escalation path for launch issues defined
+
+### Marketing
+- [x] Landing page «AI Summarization» published (staging, to go live June 30)
+- [x] Social media posts scheduled (LinkedIn Tuesday 9AM, Twitter 10AM)
+- [x] PR outreach: TechCrunch exclusive placed (briefing June 27)
+- [x] Customer advisory board preview (email June 25)
+
+---
+
+## 🚀 Launch Day (June 30)
+
+### Morning of launch (0700 PT)
+- [x] Final health check — all metrics green, no overnight issues
+- [x] Support team on-call (5 team members standby, dedicated Slack channel)
+- [x] PM monitoring dashboard (first 2 hours intensive, 30 min check-ins thereafter)
+- [x] Feature flag sequence:
+  - 0900 PT: Enable for AI Tier accounts (100% of tier — 40 accounts)
+  - 0930 PT: Expose upsell banner to Core tier customers
+
+### Communications (0900 PT)
+- [x] Email sent to customer list (4200 customer contacts, 4 variants by segment + tier)
+- [x] In-app banner activated
+- [x] Blog post published
+- [x] Changelog updated live
+- [x] Social media posts published (LinkedIn, Twitter)
+- [x] Sales team Slack notification («AI GA is live — use deck v3»)
+
+### Monitoring (ongoing)
+- Error rate: baseline 0.1%, launch day target <1% ✅ (actual: 0.3%)
+- p95 latency: target <60s ✅ (actual: 42s)
+- Generation volume: 1,200 summaries day 1 ✅ (within capacity)
+- Support ticket volume: target <5% of total ✅ (actual: 3.1%)
+
+---
+
+## 📊 Post-Launch Monitoring (T+1 week, June 30 - July 7)
+
+### Daily (M-F, 9am standup)
+- [x] Metrics dashboard review (PM + Data + SRE)
+- [x] Support ticket trend review (CS Lead)
+- [x] NPS feedback check (survey opens after first summary use)
+
+### End of week 1 review (July 7)
+- [x] Success metric check:
+  - AI Tier adoption: 43% of AI Tier managers (target 50% by day 30 — on track)
+  - Tier upgrade funnel: 8 new upgrades week 1 (target 15 by day 30 — on track)
+  - NPS: 52 (baseline 45, target 50 — ✅ exceeded)
+- [x] Adoption rate by segment: SMB 35%, mid-market 58%, enterprise 68%
+- [x] Issues discovered: 7 new (2 resolved week 1, 5 in backlog)
+- [x] Customer feedback themes: «Love it!» (majority), «Want mobile» (3 customers), «Privacy concerns resolved» (positive signal)
+
+---
+
+## 🔄 Post-Launch Review (T+4 weeks, July 28)
+
+- [ ] Success metrics vs targets (retrospective)
+- [ ] Adoption analysis by segment + cohort
+- [ ] Support ticket themes (quantitative)
+- [ ] Churn / expansion impact (analyst review)
+- [ ] Lessons learned document (cross-functional)
+- [ ] Roadmap adjustments based on learnings
+
+---
+
+## Rollback Plan
+
+### Trigger Conditions (ANY of):
+- **Error rate > 2% sustained for > 15 minutes** → immediate flag disable
+- **P95 latency > 120s sustained for > 30 minutes** → investigate + consider disable
+- **Critical customer escalation from Enterprise account** → review within 2 hours
+- **Security incident (data leak / unauthorized access)** → immediate disable + customer notification
+- **LLM provider SLA breach (both primary + fallback)** → automatic failover + investigate
+
+### Rollback Mechanism
+1. **Primary:** LaunchDarkly feature flag disable («ai_summarization_ga» → false)
+2. **Time to rollback:** ≤ 5 minutes (flag propagation + cache invalidation)
+3. **User impact:** AI toggle greys out, existing AI summaries remain accessible (read-only)
+4. **Data impact:** No data loss — summaries persisted, just generation paused
+
+### Communication Plan (on rollback)
+- **Internal:** Slack #incident-response + PagerDuty ack
+- **Customers:** In-app banner «AI Summarization temporarily paused — investigating» (within 10 min)
+- **Affected customer call-out:** Email to Enterprise customers within 1 hour
+- **Status page:** Public status.teamflow.com updated
+
+### Post-Rollback
+- Root cause analysis within 48 hours (5-why analysis)
+- Fix verified in staging + beta customers
+- Re-launch plan with learnings (may be next day, next week, or delay)
+
+---
+
+## Ownership Per Section
+
+| Section | Owner | Backup |
+|---------|-------|--------|
+| Engineering readiness | Priya (Eng Lead) | Marcus (Backend) |
+| Product / feature correctness | Alex (PM) | Jordan (Design) |
+| Legal / Compliance | Mia (Legal) | Sarah (Security) |
+| Sales enablement | VP Sales | Sales Ops |
+| CS / Support | CS Lead | Support Manager |
+| Marketing / Customer Comms | Marketing Lead | CMO |
+| Launch Day coordination | Jordan (Launch Commander) | Alex (PM backup) |
+| Rollback decision authority | Priya (Eng Lead) + Alex (PM), can decide alone in emergency | SRE on-call |
+```
+
+> **launch-checklist lesson:** A checklist of 100+ items is normal for a major B2B SaaS launch. **Rollback plan with numeric triggers** — «error rate >2% for >15 min» — not «if it breaks». **Ownership matrix** prevents «no one noticed» — each section has primary + backup. **Phased feature flag rollout** (Tier customers first, then upsell to Core) minimizes blast radius. **Communications across multi-channel** (email + in-app + blog + social + PR + sales enablement) — a single channel never reaches all stakeholders.

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

@@ -0,0 +1,15 @@
+{
+  "name": "launch-checklist",
+  "display_name": "Launch Checklist",
+  "description": "Launch checklist — internal (dogfooding, training, rollback) + external (GTM, customer comms)",
+  "default_prompt": "Use $launch-checklist when the task relates to the skill \"Launch Checklist\".",
+  "triggers": [
+    "launch-checklist",
+    "launch checklist"
+  ],
+  "capabilities": [
+    "launch-checklist"
+  ],
+  "tools": [],
+  "implicit_invocation": true
+}

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

@@ -0,0 +1,15 @@
+{
+  "name": "launch-checklist",
+  "display_name": "Launch Checklist",
+  "description": "Launch checklist — internal (dogfooding, training, rollback) + external (GTM, customer comms)",
+  "default_prompt": "Use $launch-checklist when the task relates to the skill \"Launch Checklist\".",
+  "triggers": [
+    "launch-checklist",
+    "launch checklist"
+  ],
+  "capabilities": [
+    "launch-checklist"
+  ],
+  "tools": [],
+  "implicit_invocation": true
+}

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

@@ -0,0 +1,15 @@
+{
+  "name": "launch-checklist",
+  "display_name": "Launch Checklist",
+  "description": "Launch checklist — internal (dogfooding, training, rollback) + external (GTM, customer comms)",
+  "default_prompt": "Use $launch-checklist when the task relates to the skill \"Launch Checklist\".",
+  "triggers": [
+    "launch-checklist",
+    "launch checklist"
+  ],
+  "capabilities": [
+    "launch-checklist"
+  ],
+  "tools": [],
+  "implicit_invocation": true
+}

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

@@ -0,0 +1,8 @@
+interface:
+  display_name: "Launch Checklist (B2B SaaS)"
+  short_description: "Launch checklist — internal (dogfooding, training, rollback) + external (GTM, customer comms)"
+  default_prompt: "Use $launch-checklist when the task relates to the skill \"Launch Checklist (B2B SaaS)\"."
+dependencies:
+  tools: []
+policy:
+  allow_implicit_invocation: true

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

@@ -0,0 +1,15 @@
+{
+  "name": "launch-checklist",
+  "display_name": "Launch Checklist",
+  "description": "Launch checklist — internal (dogfooding, training, rollback) + external (GTM, customer comms)",
+  "default_prompt": "Use $launch-checklist when the task relates to the skill \"Launch Checklist\".",
+  "triggers": [
+    "launch-checklist",
+    "launch checklist"
+  ],
+  "capabilities": [
+    "launch-checklist"
+  ],
+  "tools": [],
+  "implicit_invocation": true
+}

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

@@ -0,0 +1,18 @@
+version: 1
+name: "launch-checklist"
+display_name: "Launch Checklist"
+description: "Launch checklist — internal (dogfooding, training, rollback) + external (GTM, customer comms)"
+default_prompt: "Use $launch-checklist when the task relates to the skill \"Launch Checklist (B2B SaaS)\"."
+triggers:
+  - "launch-checklist"
+  - "launch checklist"
+capabilities:
+  - "launch-checklist"
+tools: []
+invocation:
+  explicit: true
+  implicit: true
+localization:
+  default_locale: "en"
+  available_locales:
+    - "en"