aiox-core 5.0.2 → 5.0.4

package/pro/squads/squad-creator-pro/tasks/next-squad.md

@@ -0,0 +1,487 @@
+# Task: Next Squad Recommendation
+
+**Task ID:** next-squad
+**Version:** 1.0.0
+**Execution Type:** Hybrid (Script metrics + Agent analysis)
+**Purpose:** Determine which squad should be created or improved next based on ecosystem analysis, business gaps, and backlog
+**Orchestrator:** @squad-chief
+**Mode:** Autonomous with optional elicitation
+**Model:** `Sonnet` (analysis + synthesis over structured data)
+
+**Frameworks Used:**
+- `data/squad-registry.yaml` → Ecosystem state, gaps, extension candidates (Phase 1)
+- `data/quality-dimensions-framework.md` → Scoring criteria (Phase 3)
+- `data/tier-system-framework.md` → Squad maturity classification (Phase 2)
+
+---
+
+## Overview
+
+This task answers the question: **"What squad should I build or improve next?"**
+
+It analyzes 4 data sources, scores candidates across 5 dimensions, and outputs a ranked recommendation with clear rationale. Designed to be called standalone (`*next-squad`) or as Phase 0 of `*create-squad-smart`.
+
+```
+INPUT (optional: business_context, priority_domain)
+    ↓
+[PHASE 1: ECOSYSTEM SCAN]
+    → Read squad-registry.yaml (current state)
+    → Identify extension candidates, config issues, gaps
+    → Map domain coverage vs known needs
+    ↓
+[PHASE 2: SIGNAL COLLECTION]
+    → Scan workspace for business signals (PRDs, stories, outputs)
+    → Check recent git activity for momentum indicators
+    → Detect implicit demand (frequent agent activations, failed lookups)
+    ↓
+[PHASE 3: SCORING]
+    → Score each candidate across 5 dimensions
+    → Rank by composite score
+    → Separate into CREATE vs IMPROVE buckets
+    ↓
+[PHASE 4: RECOMMENDATION]
+    → Present top 3 candidates with rationale
+    → Show effort estimate (S/M/L)
+    → Suggest execution path (*create-squad-smart or *brownfield-upgrade)
+    ↓
+OUTPUT: Ranked recommendation + execution command
+```
+
+---
+
+## Inputs
+
+| Parameter | Type | Required | Description | Example |
+|-----------|------|----------|-------------|---------|
+| `business_context` | string | No | Optional business priority or constraint | "launching SaaS product Q2" |
+| `priority_domain` | string | No | Optional domain to bias scoring toward | "sales", "finance" |
+| `mode` | enum | No | `quick` (registry only) or `deep` (full workspace scan) | Default: `quick` |
+
+---
+
+## Preconditions
+
+- [ ] `squads/squad-creator-pro/data/squad-registry.yaml` exists and is current
+- [ ] At least 1 squad exists in `squads/` directory
+
+**Recommended:** Run `*refresh-registry` before `*next-squad` for freshest data.
+
+---
+
+## PHASE 1: ECOSYSTEM SCAN
+
+### Step 1.1: Load Registry State
+
+**Action:**
+```yaml
+action: read
+file: squads/squad-creator-pro/data/squad-registry.yaml
+extract:
+  - metadata.total_squads
+  - gaps.extension_candidates      # Squads that exist but are incomplete
+  - gaps.potential_domains          # Domains not yet covered
+  - gaps.removed_since_last_scan   # Squads that were removed (possible re-creation)
+  - ecosystem_health.config_issues # Squads with structural problems
+  - summary.*                      # Totals for context
+```
+
+### Step 1.2: Build Candidate Lists
+
+**Action:** From registry data, build 3 candidate lists:
+
+```yaml
+candidates:
+  CREATE:
+    source: gaps.potential_domains + gaps.removed_since_last_scan
+    description: "Squads that don't exist but could fill a business need"
+    examples:
+      - domain: "sales"
+        reason: "No sales methodology squad (Sandler, Challenger, SPIN)"
+      - domain: "finance"
+        reason: "No financial analysis or FinOps squad"
+
+  IMPROVE:
+    source: gaps.extension_candidates
+    description: "Squads that exist but are operationally incomplete"
+    signals:
+      - agents > 0 AND tasks == 0     → "Has minds but no workflows"
+      - agents > 0 AND workflows == 0 → "Has agents but no orchestration"
+      - has_readme == false            → "Undocumented"
+      - config_issues present          → "Structural problems"
+
+  FIX:
+    source: ecosystem_health.config_issues
+    description: "Squads with broken configs or structural issues"
+    priority: "Quick wins — small effort, immediate health improvement"
+```
+
+### Step 1.3: Map Domain Coverage
+
+**Action:** Build a coverage matrix showing which business domains are served:
+
+```
+Domain Coverage Matrix:
+├── Strategy      → advisory-board ✅
+├── Operations    → c-level, hormozi ✅
+├── Marketing     → copy, storytelling, traffic-masters, movimento ✅
+├── Brand         → brand ✅
+├── Technical     → cybersecurity, db-sage, design, etl-ops ✅
+├── Data          → data, spy ✅
+├── Research      → deep-research ⚠️ (agents only, no tasks)
+├── Meta          → squad-creator ✅
+├── Sales         → ❌ GAP
+├── Finance       → ❌ GAP
+├── Product       → ❌ GAP
+├── Legal         → ❌ GAP (was removed)
+├── HR            → ❌ GAP (was removed)
+├── Education     → ❌ GAP
+└── Health        → ❌ GAP
+```
+
+---
+
+## PHASE 2: SIGNAL COLLECTION
+
+> **Only in `deep` mode.** Skip to Phase 3 if `mode=quick`.
+
+### Step 2.1: Workspace Activity Scan
+
+**Action:** Check recent workspace activity for demand signals:
+
+```yaml
+scan_locations:
+  - pattern: "docs/stories/**/*.md"
+    look_for: "References to domains without squads"
+    signal: "Story mentions 'sales funnel' but no sales squad → demand signal"
+
+  - pattern: "docs/prd/**/*.md"
+    look_for: "PRDs that imply new domain needs"
+    signal: "PRD for 'financial dashboard' → finance squad demand"
+
+  - pattern: "outputs/**/*"
+    look_for: "Ad-hoc outputs in domains without squads"
+    signal: "outputs/legal-analysis/ exists but no legal squad → organic demand"
+
+  - pattern: ".aios/project-status.yaml"
+    look_for: "Current epic/story context"
+    signal: "Active work in a domain strengthens that domain's score"
+```
+
+### Step 2.2: Git Momentum Analysis
+
+**Action:** Check recent commits for domain activity:
+
+```bash
+# Last 20 commits - which squads are being actively developed?
+git log --oneline -20 --name-only | grep "squads/"
+```
+
+**Interpretation:**
+- Squads with recent commits → Active development, may need supporting squads
+- Squads with NO recent commits → Stale, may need revival or removal
+- Non-squad commits mentioning domain keywords → Implicit demand
+
+### Step 2.3: Removed Squad Analysis
+
+**Action:** Cross-reference `gaps.removed_since_last_scan` with current needs:
+
+```yaml
+removed_squads_triage:
+  for_each: gaps.removed_since_last_scan
+  check:
+    - "Was it removed intentionally (deprecated) or accidentally (cleanup)?"
+    - "Does the domain still have business relevance?"
+    - "Are there residual artifacts in outputs/ or docs/?"
+  classify:
+    - REVIVE: "Domain still relevant, removal was cleanup, re-create with current standards"
+    - ARCHIVE: "Domain no longer relevant, document why, move on"
+    - MERGE: "Functionality absorbed by another squad"
+```
+
+---
+
+## PHASE 3: SCORING
+
+### Step 3.1: Score Each Candidate
+
+**Action:** Apply 5-dimension scoring (1-5 each, max 25):
+
+```yaml
+scoring_dimensions:
+  D1_business_impact:
+    question: "How much does this squad accelerate business goals?"
+    weight: 1.5  # Most important
+    scoring:
+      5: "Directly unblocks revenue or core product"
+      4: "Supports active business initiative"
+      3: "General operational improvement"
+      2: "Nice to have, no urgency"
+      1: "Theoretical benefit only"
+
+  D2_ecosystem_synergy:
+    question: "How well does this squad integrate with existing squads?"
+    weight: 1.2
+    scoring:
+      5: "Fills critical gap between 2+ active squads"
+      4: "Complements active squad workflows"
+      3: "Independent but compatible"
+      2: "Minimal overlap or connection"
+      1: "Isolated domain"
+
+  D3_effort_vs_reward:
+    question: "What's the ROI of creating/improving this squad?"
+    weight: 1.0
+    scoring:
+      5: "Low effort, high reward (FIX/quick extension)"
+      4: "Medium effort, high reward"
+      3: "Medium effort, medium reward"
+      2: "High effort, medium reward"
+      1: "High effort, low reward"
+
+  D4_demand_signals:
+    question: "How many signals indicate this squad is needed?"
+    weight: 1.3
+    scoring:
+      5: "3+ explicit signals (PRD, stories, user requests)"
+      4: "2 signals (workspace activity + business need)"
+      3: "1 clear signal"
+      2: "Inferred from ecosystem gaps only"
+      1: "No signals, purely speculative"
+
+  D5_foundation_readiness:
+    question: "How ready is the foundation to build on?"
+    weight: 1.0
+    scoring:
+      5: "Agents exist, just need tasks/workflows (IMPROVE)"
+      4: "Minds already researched, sources available"
+      3: "Domain has well-known elite minds"
+      2: "Domain exists but elite minds unclear"
+      1: "Niche domain, research-heavy"
+```
+
+### Step 3.2: Calculate Composite Score
+
+**Formula:**
+```
+composite = (D1 × 1.5) + (D2 × 1.2) + (D3 × 1.0) + (D4 × 1.3) + (D5 × 1.0)
+max_possible = 25 × avg_weight = 30.0
+normalized = (composite / 30.0) × 10  # Scale 0-10
+```
+
+### Step 3.3: Rank and Classify
+
+**Action:** Sort candidates by composite score, classify by action type:
+
+```yaml
+classification:
+  score >= 7.0: "🔴 HIGH PRIORITY — Do next"
+  score 5.0-6.9: "🟡 MEDIUM — Plan for soon"
+  score < 5.0: "🟢 LOW — Backlog"
+
+action_type:
+  FIX: "Config fix, README addition, quick structural repair"
+  IMPROVE: "Add tasks/workflows to existing squad"
+  CREATE: "Full squad creation from research"
+```
+
+---
+
+## PHASE 4: RECOMMENDATION
+
+### Step 4.1: Present Top 3
+
+**Output format:**
+
+```markdown
+## 🎯 Next Squad Recommendation
+
+### Analysis Summary
+- **Ecosystem:** {total_squads} squads | {total_agents} agents | {total_tasks} tasks
+- **Domain Coverage:** {covered}/{total_domains} domains covered
+- **Health Issues:** {config_issues_count} squads with config problems
+- **Extension Candidates:** {extension_count} squads need completion
+
+---
+
+### #1: {CANDIDATE_NAME} — {ACTION_TYPE}
+**Score:** {normalized}/10 | **Effort:** {S|M|L} | **Type:** {CREATE|IMPROVE|FIX}
+
+| Dimension | Score | Rationale |
+|-----------|-------|-----------|
+| Business Impact | {D1}/5 | {reason} |
+| Ecosystem Synergy | {D2}/5 | {reason} |
+| Effort vs Reward | {D3}/5 | {reason} |
+| Demand Signals | {D4}/5 | {reason} |
+| Foundation Ready | {D5}/5 | {reason} |
+
+**Why this is #1:** {2-3 sentence explanation}
+**Execute:** `*{command} {args}`
+
+---
+
+### #2: {CANDIDATE_NAME} — {ACTION_TYPE}
+...
+
+### #3: {CANDIDATE_NAME} — {ACTION_TYPE}
+...
+
+---
+
+### Quick Wins (FIX — do anytime)
+- [ ] {squad}: {issue} → {fix command}
+- [ ] {squad}: {issue} → {fix command}
+```
+
+### Step 4.2: Suggest Execution Path
+
+**Decision tree:**
+
+```yaml
+execution_path:
+  if_action == FIX:
+    command: "Direct fix (no workflow needed)"
+    example: "Fix cybersecurity config.yaml syntax error"
+    effort: S
+
+  if_action == IMPROVE:
+    command: "*brownfield-upgrade {squad_name}"
+    example: "Add tasks and workflows to deep-research squad"
+    effort: M
+
+  if_action == CREATE:
+    command: "*create-squad-smart {domain}"
+    example: "Create sales squad from scratch"
+    effort: L
+```
+
+---
+
+## Outputs
+
+| Output | Format | Location |
+|--------|--------|----------|
+| Recommendation report | Markdown | Displayed in terminal |
+| Candidate scores | YAML (optional) | `outputs/next-squad/scores-{date}.yaml` |
+
+**Primary output:** Terminal display with actionable recommendation.
+**Optional:** Save scores to file with `--save` flag.
+
+---
+
+## Validation
+
+```yaml
+validation:
+  - criterion: "At least 3 candidates identified and scored"
+    type: blocking
+
+  - criterion: "All 5 dimensions scored for top 3 candidates"
+    type: blocking
+
+  - criterion: "Each recommendation includes executable command"
+    type: blocking
+
+  - criterion: "Scores are justified with evidence, not assumptions"
+    type: blocking
+
+  - criterion: "business_context (if provided) influenced scoring"
+    type: warning
+
+  - criterion: "Registry data is < 24h old"
+    type: warning
+    fix: "Run *refresh-registry first"
+```
+
+---
+
+## Veto Conditions
+
+```yaml
+veto:
+  - condition: "squad-registry.yaml not found or empty"
+    action: "BLOCK — run *refresh-registry first"
+
+  - condition: "All candidates score < 3.0"
+    action: "WARN — ecosystem may be mature, consider domain expansion research"
+
+  - condition: "business_context contradicts top recommendation"
+    action: "RERANK — user priority overrides composite score"
+```
+
+---
+
+## Command Integration
+
+```bash
+# Quick mode (registry analysis only)
+*next-squad
+
+# Deep mode (full workspace scan)
+*next-squad --deep
+
+# With business context
+*next-squad --context "launching SaaS product"
+
+# Biased toward specific domain
+*next-squad --domain sales
+
+# Save scores to file
+*next-squad --save
+```
+
+---
+
+## Examples
+
+### Example 1: Quick Mode Output
+
+```
+🎯 Next Squad Recommendation
+
+Analysis: 17 squads | 164 agents | 423 tasks | 12/19 domains covered
+
+#1: deep-research — IMPROVE
+Score: 8.2/10 | Effort: M | Type: IMPROVE
+Why: 11 elite minds (Kahneman, Cochrane) already cloned but ZERO tasks/workflows.
+     Highest ROI — agents exist, just need operational artifacts.
+Execute: *brownfield-upgrade deep-research
+
+#2: sales — CREATE
+Score: 7.1/10 | Effort: L | Type: CREATE
+Why: Zero coverage for sales methodology. Synergy with copy + hormozi squads.
+     Well-known elite minds available (Sandler, Challenger, SPIN Selling).
+Execute: *create-squad-smart sales
+
+#3: cybersecurity — FIX
+Score: 6.8/10 | Effort: S | Type: FIX
+Why: config.yaml has YAML syntax error. Quick fix, immediate health gain.
+Execute: Fix squads/cybersecurity/config.yaml line 422-423
+
+Quick Wins:
+- [ ] cybersecurity: YAML syntax error → fix config.yaml
+- [ ] c-level: Missing config.yaml → generate from template
+- [ ] deep-research: Missing README + config → generate basics
+```
+
+### Example 2: Deep Mode with Context
+
+```
+*next-squad --deep --context "preparing for product launch"
+
+🎯 Next Squad Recommendation (Deep Analysis)
+
+Analysis: 17 squads | 164 agents | Workspace signals: 4 detected
+
+#1: sales — CREATE
+Score: 8.7/10 | Effort: L | Type: CREATE
+Why: Product launch demands sales methodology. Found 2 PRD references to
+     "sales funnel" and "conversion optimization" without squad coverage.
+     Direct synergy with copy (persuasion) + hormozi (offers) + traffic-masters (leads).
+Execute: *create-squad-smart sales
+```
+
+---
+
+_Task Version: 1.0.0_
+_Created: 2026-02-18_
+_Author: squad-chief_