@dzhechkov/keysarium 1.1.2 → 1.3.0

package/templates/.claude/skills/feature-adr/references/adr-template.md

@@ -0,0 +1,102 @@
+# ADR Template
+
+Use this template for every Architecture Decision Record.
+
+---
+
+```markdown
+# ADR-{NNN}: {Short Title of Decision}
+
+## Status
+
+{Proposed | Accepted | Deprecated | Superseded by ADR-{M}}
+
+## Date
+
+{YYYY-MM-DD}
+
+## Context
+
+{What is the issue that we're seeing that is motivating this decision or change?
+Describe the forces at play — technical, business, political, social.
+Describe the problem space, not the solution.}
+
+## Decision Drivers
+
+- {driver 1: e.g., "Team has no experience with technology X"}
+- {driver 2: e.g., "Feature requires sub-100ms latency"}
+- {driver 3: e.g., "Must integrate with existing auth system"}
+
+## Considered Options
+
+### Option 1: {Name}
+
+{1-2 paragraph description of this option}
+
+- ✅ {Pro 1}
+- ✅ {Pro 2}
+- ❌ {Con 1}
+- ❌ {Con 2}
+
+### Option 2: {Name}
+
+{1-2 paragraph description}
+
+- ✅ {Pro 1}
+- ❌ {Con 1}
+
+### Option 3: {Name} (optional)
+
+{1-2 paragraph description}
+
+- ✅ {Pro 1}
+- ❌ {Con 1}
+
+## Decision Matrix
+
+| Criterion | Weight | Option 1 | Option 2 | Option 3 |
+|-----------|--------|----------|----------|----------|
+| {Criterion 1} | {1-5} | {1-5} | {1-5} | {1-5} |
+| {Criterion 2} | {1-5} | {1-5} | {1-5} | {1-5} |
+| {Criterion 3} | {1-5} | {1-5} | {1-5} | {1-5} |
+| **Weighted Total** | | **{sum}** | **{sum}** | **{sum}** |
+
+## Decision
+
+We chose **{Option N}** because {1-2 sentence rationale linking back to decision drivers}.
+
+## Consequences
+
+### Positive
+- {Positive consequence 1}
+- {Positive consequence 2}
+
+### Negative
+- {Negative consequence 1 — with mitigation strategy}
+
+### Risks
+- {Risk 1} — Mitigation: {how to mitigate}
+- {Risk 2} — Mitigation: {how to mitigate}
+
+## Links
+
+- Requirement: FR-{N}
+- Related: ADR-{M} (if any)
+```
+
+---
+
+## ADR Numbering
+
+- Start from 001 for each feature
+- Use kebab-case slugs: `001-choose-database-engine.md`
+- Keep sequential within a feature, no gaps
+
+## ADR Quality Checklist
+
+- [ ] At least 2 options considered
+- [ ] Every option has both pros and cons
+- [ ] Decision matrix used for non-trivial decisions
+- [ ] Consequences include both positive AND negative
+- [ ] Risks have mitigation strategies
+- [ ] Links to requirements present

package/templates/.claude/skills/feature-adr/references/c4-template.md

@@ -0,0 +1,120 @@
+# C4 Diagram Templates
+
+Reference templates for C4 architecture diagrams in Mermaid format.
+
+---
+
+## Level 1: System Context
+
+```mermaid
+graph TB
+    %% Actors
+    User["👤 User<br/><i>Description of user</i>"]
+    Admin["👤 Admin<br/><i>Description of admin</i>"]
+
+    %% System
+    System["🖥️ Our System<br/><i>What it does</i>"]
+
+    %% External Systems
+    ExtAuth["🔐 Auth Provider<br/><i>OAuth 2.0</i>"]
+    ExtPay["💳 Payment Gateway<br/><i>Stripe/etc</i>"]
+    ExtEmail["📧 Email Service<br/><i>SendGrid/etc</i>"]
+
+    %% Relationships
+    User -->|"Uses"| System
+    Admin -->|"Manages"| System
+    System -->|"Authenticates via"| ExtAuth
+    System -->|"Processes payments via"| ExtPay
+    System -->|"Sends emails via"| ExtEmail
+
+    %% Styling
+    classDef system fill:#1168bd,stroke:#0b4884,color:#fff
+    classDef external fill:#999,stroke:#666,color:#fff
+    classDef person fill:#08427b,stroke:#052e56,color:#fff
+    class System system
+    class ExtAuth,ExtPay,ExtEmail external
+    class User,Admin person
+```
+
+## Level 2: Container
+
+```mermaid
+graph TB
+    subgraph boundary ["System Boundary"]
+        WebApp["🌐 Web Application<br/><i>React / Next.js</i>"]
+        API["⚙️ API Server<br/><i>Node.js / Express</i>"]
+        Worker["🔄 Background Worker<br/><i>Bull / Agenda</i>"]
+        DB[("💾 Database<br/><i>PostgreSQL</i>")]
+        Cache[("⚡ Cache<br/><i>Redis</i>")]
+        Queue["📬 Message Queue<br/><i>Redis / RabbitMQ</i>"]
+    end
+
+    User["👤 User"] -->|"HTTPS"| WebApp
+    WebApp -->|"REST/GraphQL"| API
+    API -->|"SQL"| DB
+    API -->|"Read/Write"| Cache
+    API -->|"Publish"| Queue
+    Queue -->|"Consume"| Worker
+    Worker -->|"SQL"| DB
+```
+
+## Level 3: Component
+
+```mermaid
+graph TB
+    subgraph container ["API Server"]
+        Router["🔀 Router<br/><i>Express Router</i>"]
+        AuthMW["🔐 Auth Middleware<br/><i>JWT validation</i>"]
+        Controller["📋 Controller<br/><i>Request handling</i>"]
+        Validator["✅ Validator<br/><i>Input validation</i>"]
+        Service["⚙️ Service<br/><i>Business logic</i>"]
+        Repository["💾 Repository<br/><i>Data access</i>"]
+        EventBus["📡 Event Bus<br/><i>Domain events</i>"]
+    end
+
+    Router --> AuthMW
+    AuthMW --> Controller
+    Controller --> Validator
+    Controller --> Service
+    Service --> Repository
+    Service --> EventBus
+    Repository --> DB[("Database")]
+    EventBus --> Queue["Message Queue"]
+```
+
+## Sequence Diagram Template
+
+```mermaid
+sequenceDiagram
+    actor User
+    participant FE as Frontend
+    participant API as API Server
+    participant Auth as Auth Service
+    participant DB as Database
+    participant Queue as Message Queue
+
+    User->>FE: Action
+    FE->>API: POST /api/resource
+    API->>Auth: Validate token
+    Auth-->>API: Valid
+
+    API->>API: Validate input
+    API->>DB: INSERT INTO resource
+    DB-->>API: Created
+
+    API->>Queue: Publish event
+    API-->>FE: 201 Created
+    FE-->>User: Success feedback
+
+    Note over Queue: Async processing
+    Queue->>Worker: Process event
+    Worker->>DB: UPDATE status
+```
+
+## Tips
+
+1. **Keep diagrams focused** — one concept per diagram
+2. **Use consistent naming** — match code namespaces
+3. **Show data flow direction** — arrows point from caller to callee
+4. **Label relationships** — what protocol/format
+5. **Highlight the new** — mark new components vs existing

package/templates/.claude/skills/feature-adr/references/complexity-matrix.md

@@ -0,0 +1,139 @@
+# Complexity Matrix — Feature Classification Reference
+
+## Dimension Scoring Guide
+
+### 1. Files Affected
+
+| Score | Range | Examples |
+|-------|-------|---------|
+| 1 (S) | 1-3 files | Fix typo, update config, add util function |
+| 2 (M) | 4-10 files | New endpoint + tests + migration + handler |
+| 3 (L) | 11-30 files | New module with models, services, controllers, tests |
+| 4 (XL) | 30+ files | Framework migration, new subsystem |
+
+**How to estimate:** Count modified + new + deleted files. Include tests.
+
+### 2. Domains Touched
+
+| Score | Range | Examples |
+|-------|-------|---------|
+| 1 (S) | 1 domain | Change within a single bounded context |
+| 2 (M) | 1-2 domains | Feature spanning API + database |
+| 3 (L) | 2-4 domains | Feature touching auth + billing + notifications |
+| 4 (XL) | 4+ domains | Cross-cutting concern affecting all domains |
+
+**How to identify domains:** Bounded contexts, microservices, top-level modules.
+
+### 3. New Integrations
+
+| Score | Range | Examples |
+|-------|-------|---------|
+| 1 (S) | 0 | No new external dependencies |
+| 2 (M) | 0-1 | Adding a new npm package, calling a new API |
+| 3 (L) | 1-3 | New database, message queue, 3rd-party service |
+| 4 (XL) | 3+ | Multiple new infrastructure components |
+
+### 4. Breaking Changes
+
+| Score | Range | Examples |
+|-------|-------|---------|
+| 1 (S) | 0 | Fully backward compatible |
+| 2 (M) | 0 | Backward compatible but behavior changes |
+| 3 (L) | 0-2 | API version bump, schema migration with downtime |
+| 4 (XL) | 2+ | Major API redesign, data model overhaul |
+
+### 5. New Data Models
+
+| Score | Range | Examples |
+|-------|-------|---------|
+| 1 (S) | 0 | No schema changes |
+| 2 (M) | 0-1 | Add field to existing model, new simple entity |
+| 3 (L) | 1-3 | New aggregate root, new relationship graph |
+| 4 (XL) | 3+ | New bounded context with full entity graph |
+
+### 6. Cross-Cutting Concerns
+
+| Score | Range | Concerns |
+|-------|-------|----------|
+| 1 (S) | 0 | None |
+| 2 (M) | 0-1 | Logging OR error handling |
+| 3 (L) | 1-2 | Auth + caching, i18n + validation |
+| 4 (XL) | 3+ | Auth + logging + caching + monitoring + i18n |
+
+---
+
+## Step Activation by Tier
+
+### Tier S (Score 6-8)
+
+```
+Active:  0 → 1(light) → 6(inline) → 7 → 8(smoke)
+Skipped: 2, 3, 4, 5
+```
+
+**Step adaptations:**
+- Step 1: 3-5 bullet requirements, no formal document
+- Step 6: Inline in conversation, no separate file
+- Step 8: Smoke tests only — does it compile? Do existing tests pass?
+
+### Tier M (Score 9-13)
+
+```
+Active:  0 → 1 → 3(1 ADR) → 5(light) → 6 → 7 → 8
+Skipped: 2, 4
+```
+
+**Step adaptations:**
+- Step 3: Single ADR for the main architectural decision
+- Step 5: Component diagram only, no full C4
+
+### Tier L (Score 14-19)
+
+```
+Active:  0 → 1 → [2 ‖ 3] → [4 ‖ 5] → 6 → 7 → 8
+Parallel groups: (2,3) and (4,5)
+```
+
+**Step adaptations:**
+- Step 2: Research analogues in codebase and external patterns
+- Step 3: Multiple ADRs for each significant decision
+- Step 4: Bounded contexts + ubiquitous language
+- Step 5: Full C4 (Context + Container + Component)
+- Step 8: Unit + integration tests + code review
+
+### Tier XL (Score 20-24)
+
+```
+Active:  0 → 1 → [2 ‖ 3] → [4 ‖ 5] → 6 → 7(parallel) → 8(full)
+Parallel groups: (2,3), (4,5), (7 per module)
+```
+
+**Step adaptations:**
+- All steps at full depth
+- Step 7: Multiple parallel agents, one per module/domain
+- Step 8: Full QE — unit + integration + e2e + performance + security review
+
+---
+
+## Quick Reference Decision Tree
+
+```
+Is it a bug fix or config change?
+  → YES → S
+
+Does it add a new feature with new files?
+  → YES, < 10 files → M
+  → YES, 10-30 files → L
+  → YES, 30+ files → XL
+
+Does it touch multiple domains?
+  → NO → max M
+  → 2-4 → min L
+  → 4+ → XL
+
+Does it have breaking changes?
+  → YES → min M, likely L/XL
+
+Does it require new infrastructure?
+  → YES → min L
+```

package/templates/.claude/skills/feature-adr/references/qe-checklist.md

@@ -0,0 +1,86 @@
+# QE Checklist
+
+Unified quality engineering checklist adapted from shift-left testing and brutal-honesty review patterns.
+
+---
+
+## Pre-Implementation (Shift-Left)
+
+Before writing code, verify:
+
+- [ ] Requirements have testable acceptance criteria
+- [ ] Edge cases identified in requirements phase
+- [ ] Data model validated against expected queries
+- [ ] API contract defined before implementation
+- [ ] Error scenarios documented
+
+## During Implementation
+
+While coding, continuously check:
+
+- [ ] Each function has a clear single responsibility
+- [ ] Error paths return meaningful messages
+- [ ] Input validation at system boundaries
+- [ ] No hardcoded secrets, URLs, or environment-specific values
+- [ ] Logging at appropriate levels (debug/info/warn/error)
+
+## Post-Implementation
+
+### Correctness Checks
+- [ ] All acceptance criteria have corresponding tests
+- [ ] Happy path tested end-to-end
+- [ ] Edge cases tested (null, empty, max, min, boundary)
+- [ ] Error paths tested (invalid input, network failure, timeout)
+- [ ] Concurrent access tested (if applicable)
+
+### Security Checks (OWASP-aligned)
+- [ ] No SQL/NoSQL injection vectors
+- [ ] No XSS vectors in output
+- [ ] Authentication required on all protected endpoints
+- [ ] Authorization checked (not just authentication)
+- [ ] Sensitive data not logged or exposed in errors
+- [ ] CSRF protection on state-changing operations
+- [ ] Rate limiting on public endpoints
+- [ ] Input size limits enforced
+
+### Performance Checks
+- [ ] No N+1 query patterns
+- [ ] Database queries use appropriate indexes
+- [ ] Large result sets are paginated
+- [ ] Expensive operations are async/cached where appropriate
+- [ ] No unnecessary memory allocations in hot paths
+- [ ] Connection pools sized appropriately
+
+### Maintainability Checks
+- [ ] Code follows existing project conventions
+- [ ] Naming is consistent with codebase
+- [ ] No dead code or commented-out blocks
+- [ ] Complex logic has explanatory comments
+- [ ] Public APIs have documentation
+- [ ] No circular dependencies introduced
+
+### Compatibility Checks
+- [ ] Backward compatible with existing clients (or migration path documented)
+- [ ] Database migration is reversible
+- [ ] Feature flag available for gradual rollout (if applicable)
+- [ ] Existing tests still pass
+
+## Review Severity Guide
+
+| Severity | Symbol | Meaning | Action |
+|----------|--------|---------|--------|
+| Blocker | 🔴 | Prevents merge | Must fix |
+| Warning | 🟡 | Should fix | Fix before next release |
+| Suggestion | 🔵 | Nice to have | Consider for future |
+| Praise | 🟢 | Well done | No action needed |
+
+## Tier-Specific Depth
+
+| Check Category | S | M | L | XL |
+|---------------|---|---|---|-----|
+| Pre-implementation | - | Partial | Full | Full |
+| Correctness | Smoke | Unit | Unit+Integ | Full |
+| Security | - | Basic | Full | Full+Pentest |
+| Performance | - | - | Basic | Full+Load |
+| Maintainability | - | Basic | Full | Full |
+| Compatibility | Smoke | Basic | Full | Full |

package/templates/CLAUDE.md

@@ -14,10 +14,26 @@
 /brain-export [путь или "all"]      — Экспорт знаний в portable JSON
 /brain-import [путь к JSON]         — Импорт знаний из другого проекта
 
+/init-platform --platform <name>    — Генерация конфигов для Cursor/OpenCode/Copilot
+
+/workers start <type>              — Запуск фонового воркера (consolidate, export-brain, health-check, dream-cycle)
+/workers status                    — Статус активных воркеров
+/workers stop <id>                 — Остановка воркера
+/workers list                      — Список доступных типов воркеров
+
+/dream run                         — Запуск dream cycle (анализ паттернов)
+/dream insights                    — Показать последние инсайты
+/dream status                      — Состояние триггеров dream cycle
+/dream clear                       — Очистить старые инсайты (оставить 10)
+
+/feature-adr [описание фичи]       — Adaptive Feature Development (9 шагов, S/M/L/XL)
+
 /bto [путь или описание]           — Build-Test-Optimize полный цикл
 /bto-build [описание]              — Генерация нового скилла/команды
 /bto-test [путь]                   — Мульти-агентная оценка артефакта
 /bto-optimize [путь]               — Эволюционная оптимизация промптов
+
+/learning-stats [--domain X] [--phase N] — Reward learning аналитика
 ```
 
 ## Структура проекта
@@ -44,7 +60,12 @@ product-keysarium-2026/
 │   │   ├── bto.md                  ← /bto — Build-Test-Optimize pipeline
 │   │   ├── bto-build.md            ← /bto-build — генерация артефактов
 │   │   ├── bto-test.md             ← /bto-test — мульти-агентная оценка
-│   │   └── bto-optimize.md         ← /bto-optimize — оптимизация промптов
+│   │   ├── bto-optimize.md         ← /bto-optimize — оптимизация промптов
+│   │   ├── feature-adr.md          ← /feature-adr — adaptive feature development
+│   │   ├── init-platform.md        ← /init-platform — мульти-платформенная генерация
+│   │   ├── learning-stats.md       ← /learning-stats — reward learning аналитика
+│   │   ├── workers.md              ← /workers — управление фоновыми воркерами
+│   │   └── dream.md                ← /dream — dream cycles (анализ паттернов)
 │   ├── shards/                     ← Governance shards (per-phase rules)
 │   │   ├── phase-0-discovery.shard.md
 │   │   ├── phase-1-explore.shard.md
@@ -53,7 +74,8 @@ product-keysarium-2026/
 │   │   ├── phase-3-solve.shard.md
 │   │   ├── phase-4-architecture.shard.md
 │   │   ├── phase-5-presentation.shard.md
-│   │   └── bto-evaluation.shard.md
+│   │   ├── bto-evaluation.shard.md
+│   │   └── feature-adr.shard.md
 │   └── skills/                     ← Навыки (building blocks)
 │       ├── bto/                    ← Build-Test-Optimize (мульти-агентная оценка)
 │       ├── explore/                ← Адаптивная кларификация задач
@@ -61,7 +83,36 @@ product-keysarium-2026/
 │       ├── goap-research-ed25519/  ← GOAP research + crypto verification
 │       ├── presentation-storyteller/ ← Презентации со storytelling
 │       ├── problem-solver-enhanced/  ← Решение проблем (TRIZ + Game Theory)
-│       └── reverse-engineering-unicorn/ ← Reverse engineering компаний
+│       ├── reverse-engineering-unicorn/ ← Reverse engineering компаний
+│       └── feature-adr/            ← Adaptive feature development (9 steps, S/M/L/XL)
+├── lib/                            ← Библиотечные модули
+│   ├── background-workers.md       ← Протокол фоновых воркеров
+│   ├── memory-protocol.md          ← Протокол memory_query / memory_store
+│   ├── reward-tracker.md           ← Аналитика и обнаружение паттернов
+│   ├── dream-engine.md             ← DreamEngine: concept graph + insight generation
+│   ├── platform-adapters.md        ← Реестр платформенных адаптеров
+│   ├── platform-templates/         ← Шаблоны для генерации конфигов
+│   │   ├── cursor.md               ← Cursor (.cursorrules)
+│   │   ├── opencode.md             ← OpenCode (.opencode/)
+│   │   └── copilot.md              ← GitHub Copilot (.github/copilot-instructions.md)
+│   └── worker-templates/           ← Шаблоны инструкций для воркеров
+│       ├── consolidate.md           ← Консолидация паттернов
+│       ├── health-check.md          ← Проверка здоровья системы
+│       ├── brain-export.md          ← Фоновый экспорт brain
+│       └── dream-cycle.md           ← Dream cycle воркер (concept graph + insights)
+├── packages/                       ← npm пакеты
+│   ├── dz-keysarium/              ← @dzhechkov/keysarium (research pipeline)
+│   ├── dz-skills-bto/             ← @dzhechkov/skills-bto (BTO evaluation)
+│   ├── dz-feature-adr/            ← @dzhechkov/skills-feature-adr (feature development)
+│   └── dz-keysarium-core/         ← @dzhechkov/keysarium-core (shared framework)
+│       ├── governance/             ← Constitution + shards + checkpoints
+│       ├── memory/                 ← Learning protocol + reward + dream engine
+│       ├── orchestration/          ← Queen protocol + topologies + workers + routing
+│       ├── verification/           ← Witness chain + judge attestation + audit trail
+│       ├── trust-tiers/            ← 4-tier classification + promotion
+│       └── platform/               ← Multi-platform adapters (Cursor, OpenCode, Copilot)
+├── features/                       ← Директории фич (/feature-adr output)
+│   └── [feature-slug]/             ← Каждая фича изолирована
 └── researches/                     ← Директории исследований
     └── [case-slug]/                ← Каждое исследование изолировано
 ```
@@ -138,15 +189,51 @@ DISCOVERY → EXPLORE  → RESEARCH → CJM PROTO    → SOLVE    → ARCHITECTU
 6 именованных loops формализуют передачу данных между фазами.
 См. `.claude/rules/feedback-loops.md` для Variable Registry.
 
-## Portable Brain
+## Portable Brain (v1.1)
 
 ```
-/brain-export [путь или "all"]  — Экспорт знаний в JSON контейнер
-/brain-import [файл.json]       — Импорт знаний из другого проекта
+/brain-export [путь или "all"]               — Экспорт знаний в JSON контейнер (v1.1 с manifest)
+/brain-export --delta <parent-brain.json>     — Дельта-экспорт (COW, JSON Patch RFC 6902)
+/brain-import [файл.json]                     — Импорт знаний из другого проекта
 ```
 
-Brain container включает: skill metadata, domain patterns, research summaries,
-harvest patterns, pipeline metrics. Selective import с merge-not-overwrite стратегией.
+Brain container v1.1 включает: SHA-256 manifest с checksum, skill metadata, domain patterns,
+research summaries, harvest patterns, pipeline metrics, reward data.
+Selective import с merge-not-overwrite стратегией. Backward compatible с v1.0.
+
+**v1.1 features:** Integrity manifest, delta exports (COW branching), 2-tier memory index,
+HOT/WARM/COLD/PURGE record lifecycle.
+
+## keysarium-core — Shared Framework Package
+
+`@dzhechkov/keysarium-core` — домен-агностичный фреймворк для multi-agent пайплайнов. Содержит протоколы, которые используются и в Keysarium, и в BTO, и могут быть использованы любой командой для создания собственного пайплайна.
+
+### Архитектура 4 пакетов
+
+```
+            @dzhechkov/keysarium-core              ← Общий фреймворк
+                ^         ^         ^
+                |         |         |
+           peerDep    peerDep   peerDep
+                |         |         |
+@dzhechkov/keysarium  skills-bto  skills-feature-adr
+  (исследования)       (оценка)    (фичи)
+```
+
+### Модули core
+
+| Модуль | Файлы | Назначение |
+|--------|-------|-----------|
+| governance/ | 3 | Constitution, shard protocol, checkpoint/promise protocol |
+| memory/ (v1.1) | 3 | memory_query/store, reward tracking, dream engine, **2-tier index, COW branching, record lifecycle** |
+| orchestration/ | 4 | Queen protocol, 6 topologies, workers, model routing |
+| verification/ | 3 | Witness chain, judge attestation, audit trail |
+| trust-tiers/ | 2 | 4-tier classification, promotion protocol |
+| platform/ | 4 | Adapter registry + templates (Cursor, OpenCode, Copilot) |
+
+### Связь с существующими файлами
+
+Core — это cleanroom-экстракция из `lib/` и `.claude/rules/`. Оригинальные файлы остаются как Keysarium-специфичные референсы. Core-версии обобщены и не содержат Keysarium-специфичной терминологии.
 
 ## Domain Templates
 
@@ -195,6 +282,13 @@ bto (Build-Test-Optimize)
   ├── modules/build.md (генерация артефактов)
   ├── modules/test.md (мульти-агентная оценка: 3 судьи)
   └── modules/optimize.md (эволюционная оптимизация)
+
+feature-adr (Adaptive Feature Development)
+  ├── explore (кларификация требований)
+  ├── problem-solver-enhanced (trade-off анализ для ADR)
+  ├── frontend-design (UI реализация, если есть)
+  ├── modules/00-08 (9 шагов pipeline)
+  └── references/ (ADR template, C4 template, QE checklist)
 ```
 
 ## BTO — Build-Test-Optimize
@@ -224,6 +318,39 @@ npx @dzhechkov/skills-bto init    # Установить BTO в любой пр
 npx @dzhechkov/skills-bto doctor  # Проверить здоровье установки
 ```
 
+## Feature ADR — Adaptive Feature Development
+
+9-шаговый pipeline для разработки фич любой сложности с Complexity Router.
+
+```
+/feature-adr [описание фичи]  — Полный адаптивный pipeline (9 шагов)
+```
+
+### Complexity Tiers
+
+| Tier | Scope | Active Steps |
+|------|-------|-------------|
+| **S** | 1-3 файла, 1 домен | 0→1→6→7→8 |
+| **M** | 4-10 файлов, 1-2 домена | 0→1→3→5→6→7→8 |
+| **L** | 11-30 файлов, 2-4 домена | Полный pipeline с параллелизмом |
+| **XL** | 30+ файлов, cross-cutting | Полный DAG + multi-agent swarm |
+
+### Pipeline шагов
+
+| Step | Name | Tiers | Модель |
+|------|------|-------|--------|
+| 0 | Complexity Router | All | haiku |
+| 1 | Requirements | All | sonnet |
+| 2 | Research | L/XL | sonnet |
+| 3 | ADR | M+ | opus |
+| 4 | DDD | L/XL | opus |
+| 5 | Architecture | M+ | opus |
+| 6 | Implementation Plan | All | sonnet |
+| 7 | Code | All | opus |
+| 8 | QE | All | sonnet |
+
+Артефакты: `features/<feature-slug>/` (не в корне, не в researches/)
+
 ## Anti-Patterns
 
 | Anti-Pattern | Что делать |