┌─────────────────────────────────────────────────────────────┐
│  SSOT 構造                                                   │
├─────────────────────────────────────────────────────────────┤
│  docs/features/{id}/CONTEXT.json   → priority セクション (SSOT)   │
│                                       RICE 入力値, 証拠, スコア│
│  docs/features/{id}/BRIEF.md       → Reach/Impact/Effort    │
│  docs/features/{id}/SPEC.md        → Confidence/Effort      │
│  docs/analysis/competitor-registry.json → Competitive Adj.   │
│  docs/analysis/gap-candidates.json → Impact (gap severity)   │
│  docs/research/*.md                → Confidence (参照)       │
└─────────────────────────────────────────────────────────────┘

⚠️ 原則:
   - priority データは CONTEXT.json でのみ管理
   - スコア算出時に実際に読んだファイルを data_sources_read に記録
   - 分析結果はユーザー確認後に更新

Score = (Reach × Impact × Confidence) / Effort

RICE Score = (Reach × Impact × Confidence) / Effort
Adjusted Score = RICE Score × Competitive Adjustment × Manual Override

Confidence = Σ(weight_i × factor_i)   // 各 factor は 0.0-1.0 連続値
Final_Confidence = clamp(Confidence, 0.05, 1.0)

RICE Score = (Reach × Impact × Confidence) / Effort          // 純粋 RICE
Adjusted Score = RICE Score × Competitive Adjustment × Manual Override

{
  "priority": {
    "version": "3.0",
    "schema": "rice-v2",
    "last_updated": "2026-02-09T12:00:00+00:00",
    "phase": "mvp",

    "rice_inputs": {
      "reach": {
        "score": 7,
        "evidence": {
          "target_user_scope": { "value": "all learners", "source": "BRIEF.md §1" },
          "business_metrics": { "value": "MAU retention", "source": "BRIEF.md §7" },
          "progress_adjustment": { "value": 30, "source": "CONTEXT.json progress" }
        },
        "rationale": "user_scope=9, biz=7, progress_factor=5"
      },
      "impact": {
        "score": 2,
        "evidence": {
          "core_goal": { "value": "AI 基盤カスタム学習", "source": "BRIEF.md §1" },
          "ltv_contribution": { "value": "LTV 直接貢献", "source": "BRIEF.md §7" },
          "gap_severity": { "value": "HIGH", "source": "gap-candidates.json" }
        },
        "rationale": "Impact=High (2). Gap severity=HIGH"
      },
      "confidence": {
        "score": 0.47,
        "factors": {
          "brief_completeness": { "score": 0.8, "section_count": 6, "source": "BRIEF.md" },
          "spec_quality": { "score": 0.7, "fr_count": 5, "source": "SPEC.md" },
          "research_evidence": { "score": 0.4, "research_count": 2, "sources": ["R-20260110-001"] },
          "implementation_status": { "score": 0.3, "progress_pct": 30, "source": "CONTEXT.json" }
        }
      },
      "effort": {
        "score": 4.2,
        "unit": "person-weeks",
        "evidence": {
          "fr_count": { "value": 5, "source": "SPEC.md §2" },
          "target_files": { "value": 12, "source": "SPEC.md §0" },
          "constraint_count": { "value": 3, "source": "BRIEF.md §6" },
          "remaining_pct": { "value": 70, "source": "CONTEXT.json progress" }
        },
        "rationale": "FR 5個, 70% 残り, 中規模"
      }
    },

    "competitive_adjustment": {
      "adjustment": 1.13,
      "has_data": true,
      "evidence": {
        "assessment_count": {
          "value": 3,
          "contribution": 0.03,
          "source": "competitor-registry.json"
        },
        "is_industry_standard": {
          "value": true,
          "contribution": 0.05,
          "source": "gap-candidates.json"
        },
        "opportunity_score": { "value": 7.5, "contribution": 0.1, "source": "gap-candidates.json" },
        "gap_severity": { "value": "MEDIUM", "contribution": 0.05, "source": "gap-candidates.json" }
      },
      "rationale": "CompAdj=1.13"
    },

    "calculated": {
      "rice_score": 1.57,
      "adjusted_score": 1.77,
      "competitive_adjustment": 1.13,
      "manual_override_applied": 1.0,
      "formula": "Adjusted = (R×I×C)/E × CompAdj × Override",
      "component_breakdown": {
        "numerator": 6.58,
        "denominator": 4.2
      }
    },

    "data_sources_read": [
      "BRIEF.md (date: 2026-02-09)",
      "SPEC.md (date: 2026-02-09)",
      "CONTEXT.json (date: 2026-02-09)",
      "competitor-registry.json (date: 2026-02-09)",
      "gap-candidates.json (date: 2026-02-09)"
    ],

    "manual_override": {
      "value": 1.0,
      "reason": null
    },

    "ai_rationale": "R=7, I=2, C=0.47, E=4.2, CompAdj=1.13",

    "history": [
      {
        "timestamp": "2026-02-09T12:00:00+00:00",
        "actor": "ai",
        "change_summary": "rice_calculator.py v2 再計算",
        "rice_score": 1.57,
        "adjusted_score": 1.77,
        "inputs_snapshot": {
          "reach": 7,
          "impact": 2,
          "confidence": 0.47,
          "effort": 4.2,
          "competitive_adjustment": 1.13
        }
      }
    ]
  }
}

# Reach: 候補の target_users 範囲基盤
reach = user_scope_estimate                     // 1-10 スケール (キーワード推定)
if gap_severity == "HIGH" and is_industry_standard:
    reach = min(reach + 1, 10)

# Impact: ICE Impact → RICE Impact (Intercom 5段階)
impact_map = {
    ice >= 9: 3,      // Massive
    ice >= 7: 2,      // High
    ice >= 5: 1,      // Medium
    ice >= 3: 0.5,    // Low
    else:    0.25     // Minimal
}

# Confidence: 候補は基本的に低い信頼度
confidence_base = 0.3                           // 候補デフォルト値
if source_type includes "competitor_gap":
    confidence_base += 0.15                     // 競合データあり
if evidence.length >= 2:
    confidence_base += 0.10                     // 複数証拠
// 典型的な candidate Confidence: 0.30~0.55

# Effort: ICE Effort 逆変換
effort = clamp(round((11 - ice_effort) × 1.5), 1, 15)  // person-weeks

Provisional RICE = (Reach × Impact × Confidence) / Effort

# 1. CONTEXT.json ロード
context = Read("docs/features/{feature_id}/CONTEXT.json")

# 2. schema 確認
assert context["priority"].get("schema") == "rice-v2"

# 3. BRIEF.md パース (§1-§7 セクション別分析)
brief = Read("docs/features/{feature_id}/BRIEF.md")
sections = parse_brief_sections(brief)  # §1 Problem, §5 Scope, §6 Constraints, §7 Metrics

# 4. SPEC.md パース (FR/AC カウント, Target Files)
spec = Read("docs/features/{feature_id}/SPEC.md")  # なければ None
fr_count = count_fr_definitions(spec)
target_files = extract_target_files(spec)

# 5. competitor-registry.json (competitive adjustment用)
registry = Read("docs/analysis/competitor-registry.json")

# 6. gap-candidates.json (gap severity, competitive adjustment用)
gaps = Read("docs/analysis/gap-candidates.json")

## RICE コンポーネント算出

| コンポーネント |  スコア  | 主要根拠                                |
| -------------- | :------: | --------------------------------------- |
| Reach          |    7     | 全学習者対象、高いビジネスメトリクス    |
| Impact         | 2 (High) | コアワークフロー変化, gap severity HIGH |
| Confidence     |  0.575   | SPEC あり(0.7), Research 不足(0.4)      |
| Effort         |  4.2 pw  | FR 5個, 70% 残り                        |

この値は適切ですか？修正が必要であればお知らせください。

1. 純粋 RICE
   Numerator = Reach × Impact × Confidence = 7 × 2 × 0.575 = 8.05
   Denominator = Effort = 4.2
   RICE Score = 8.05 / 4.2 = 1.92

2. 調整スコア
   Competitive Adjustment = 1.05 (競合データ基盤)
   Manual Override = 1.0 (補正なし)
   Adjusted Score = 1.92 × 1.05 × 1.0 = 2.02

## 分析結果

| 項目               | 値                |
| ------------------ | ----------------- |
| **RICE Score**     | 1.92              |
| **Adjusted Score** | 2.02              |
| **Tier**           | Tier 1 (即時着手) |
| **Confidence**     | 0.575             |

CONTEXT.json にこの結果を保存しますか？

- `--apply`: 即時保存
- 修正後保存: 値をお知らせください

┌─────────────────────────────────────────────────────────────┐
│ 自動化トリガーフロー                                            │
├─────────────────────────────────────────────────────────────┤
│                                                             │
│  feature-pilot (実装完了)                                    │
│       │                                                     │
│       ▼                                                     │
│  feature-status-sync (Step 5)                               │
│       │ ← 14日+ 経過時に警告メッセージ                          │
│       ▼                                                     │
│  priority-analyzer (Step 6, 選択的)                          │
│       │ ← progress ±10% または staleness 検知時              │
│       ▼                                                     │
│  pre-quality-gate (Step 7)                                  │
│                                                             │
└─────────────────────────────────────────────────────────────┘

CLI ツール:
  python3 .quality/scripts/rice_calculator.py --all           # 全体再計算
  python3 .quality/scripts/rice_calculator.py --feature 008   # 単一再計算
  python3 .quality/scripts/rice_calculator.py --json-output   # JSON 出力

# Priority Analysis: {feature_id}

## RICE スコア要約

| 項目               | 値                |
| ------------------ | ----------------- |
| **RICE Score**     | 1.92              |
| **Adjusted Score** | 2.02              |
| **Tier**           | Tier 1 (即時着手) |
| **Confidence**     | 0.575             |

## RICE 分解

### Reach (到達範囲)

| Evidence            | 値                     | ソース                   |
| ------------------- | ---------------------- | ------------------------ |
| Target Users        | all learners (scope 8) | BRIEF.md §1              |
| Business Metrics    | MAU retention (biz 7)  | BRIEF.md §7              |
| Progress Adjustment | 30% (factor 5)         | CONTEXT.json             |
| **Reach Score**     | **7**                  | 8×0.40 + 7×0.35 + 5×0.25 |

### Impact (影響度)

| Evidence         | 値                            | ソース              |
| ---------------- | ----------------------------- | ------------------- |
| Core Goal        | AI 基盤カスタム学習 → High(2) | BRIEF.md §1         |
| LTV Contribution | 直接貢献 → floor 2            | BRIEF.md §7         |
| Gap Severity     | HIGH → floor 2                | gap-candidates.json |
| **Impact Score** | **2 (High)**                  | Intercom snap       |

### Confidence (信頼度)

| 要素           | スコア | 加重 |   寄与    | ソース                    |
| -------------- | :----: | :--: | :-------: | ------------------------- |
| BRIEF 完成度   |  0.80  | 0.30 |   0.240   | BRIEF.md (6/7 セクション) |
| SPEC 品質      |  0.70  | 0.25 |   0.175   | SPEC.md (5 FRs)           |
| Research 根拠  |  0.40  | 0.25 |   0.100   | 2 research_ids            |
| 実装実績       |  0.30  | 0.20 |   0.060   | 30% progress              |
| **Confidence** |   -    |  -   | **0.575** |                           |

### Effort (工数)

| Evidence         | 値         | ソース       |
| ---------------- | ---------- | ------------ |
| FR Count         | 5          | SPEC.md §2   |
| Target Files     | 12         | SPEC.md §0   |
| Constraints      | 3          | BRIEF.md §6  |
| Remaining        | 70%        | CONTEXT.json |
| **Effort Score** | **4.2 pw** |              |

### 計算

## AI 分析

> {ai_rationale}

## 読んだデータソース
- BRIEF.md (date: 2026-02-09)
- SPEC.md (date: 2026-02-09)
- CONTEXT.json (date: 2026-02-09)
- competitor-registry.json (date: 2026-02-09)

# Priority Ranking Report

> 分析日: 2026-02-09
> Phase: MVP
> 分析対象: 49 Features

## Tier 1: 即時着手 (Score >= 1.5)

| 順位 | Feature             | RICE |  R  |  I  |  C   | E(pw) |
| :--: | ------------------- | :--: | :-: | :-: | :--: | :---: |
|  1   | 024-ai-tutor-system | 2.34 |  8  |  2  | 0.62 |  4.2  |
|  2   | 003-ai-content      | 1.89 |  7  |  2  | 0.55 |  4.1  |

## Tier 2: 1-2週間以内 (1.0 <= Score < 1.5)

...

## Tier 3: MVP 後 (0.5 <= Score < 1.0)

...

## Tier 4: Phase 2 (Score < 0.5)

...

## 統合優先順位ランキング

### A. 既存機能 (CONTEXT.json, RICE 確定)

| Rank | ID  | Name | RICE | R   | I   | C   | E(pw) |
| ---- | --- | ---- | ---- | --- | --- | --- | ----- |

### B. 待機候補 (scan-status.json, Provisional RICE)

| Rank | Candidate ID | Name | Prov.RICE | R   | I   | C   | E(pw) | Source | Status |
| ---- | ------------ | ---- | --------- | --- | --- | --- | ----- | ------ | ------ |

### C. 推奨: 次の変換対象

| Priority | Candidate                  | 根拠                                  |
| -------- | -------------------------- | ------------------------------------- |
| 1        | contextual-paywall-trigger | RICE 2.1, HIGH gap, industry standard |
| 2        | speaking-srs               | RICE 1.8, HIGH gap, speaking gap HIGH |

# 特定 Feature 分析
"/priority-analyzer 008-monetization"
"/priority-analyzer --feature 008,015,024"

# 分析後 CONTEXT.json に保存
"/priority-analyzer 008 --apply"

# 全 Feature 分析 (保存なしで比較のみ)
"/priority-analyzer --all"

# 全 Feature + 待機候補統合ランキング
"/priority-analyzer --all --candidates"

# Phase 変更後全体再計算
"/priority-analyzer --set-phase growth --apply"

# CLI 直接実行
# python3 .quality/scripts/rice_calculator.py --all --verbose
# python3 .quality/scripts/rice_calculator.py --feature 008 --apply