diff --git a/assets/beyond-vibe-coding/ai-aided-test-first.md b/assets/beyond-vibe-coding/ai-aided-test-first.md
new file mode 100644
index 0000000..8faa2e4
--- /dev/null
+++ b/assets/beyond-vibe-coding/ai-aided-test-first.md
@@ -0,0 +1,366 @@
+# AI-Aided Test-First Development
+
+## Официальное название
+
+**AI-aided test-first development** — официальное название от Thoughtworks Technology Radar (апрель 2023)
+
+**Статус**: **TRIAL** в Thoughtworks Technology Radar — рекомендуется для экспериментов в production
+
+---
+
+## Альтернативные названия
+
+- **Test-Driven Development with AI** (Builder.io, широко распространено)
+- **AI-Enabled TDD / AI-TDD** (AgileМania, коммерческие тренинги)
+- **TDD as AI Communication Protocol** (8th Light, концептуальное)
+- **AI-assisted TDD** (общепринятое в индустрии)
+- **Test-First AI Development** (академические источники)
+
+---
+
+## Методология
+
+### Суть подхода
+
+**Инверсия традиционного AI workflow:**
+
+| Традиционный AI подход | AI-aided test-first |
+|------------------------|---------------------|
+| Prompt → AI generates code → Human writes tests | Human defines test requirements → AI generates tests → AI implements → Tests validate |
+| AI controls scope | Tests control scope |
+| Human validates after | Human specifies before |
+| Context in conversation | Context in tests |
+
+### Цикл разработки
+
+1. **Human**: пишет test description или acceptance criteria в natural language
+2. **AI**: генерирует failing test(s) на основе описания
+3. **Human**: проверяет и корректирует тесты (they define "done")
+4. **AI**: имплементирует минимальный код для прохождения тестов
+5. **Tests**: автоматически валидируют (binary pass/fail)
+6. **AI**: рефакторит при необходимости (tests as safety net)
+7. **Repeat**: для следующей фичи
+
+### Ключевые принципы
+
+- **Tests as specification** — тесты = compressed context для AI
+- **Binary validation** — clear, measurable goal (pass/fail) вместо subjective review
+- **Human architect, AI implementer** — разработчик проектирует поведение, AI печатает код
+- **Guard rails** — тесты предотвращают AI hallucinations и scope creep
+- **Fast feedback** — instant validation вместо manual code review
+
+---
+
+## Преимущества
+
+### Для разработчика
+
+- ✅ **Контроль над scope** — тесты определяют границы, AI не может "увлечься"
+- ✅ **Сохранение навыков** — developer остаётся в роли architect/designer
+- ✅ **Clear communication** — тесты = unambiguous specification для AI
+- ✅ **Reduced cognitive load** — не нужно держать весь контекст в голове
+- ✅ **Instant validation** — automated tests вместо manual review
+
+### Для AI
+
+- ✅ **Clear goal** — binary pass/fail проще чем "good code"
+- ✅ **Structured context** — test suite как compressed requirements
+- ✅ **Reduced hallucinations** — тесты ограничивают пространство решений
+- ✅ **Measurable progress** — количество passing tests
+
+### Для команды
+
+- ✅ **Consistent quality** — junior + AI align с architectural style команды
+- ✅ **Living documentation** — тесты документируют поведение
+- ✅ **Faster onboarding** — AI помогает новичкам писать код в team's style
+- ✅ **Reduced tech debt** — TDD предотвращает "vibe coding"
+
+---
+
+## Какую проблему решает
+
+### Проблемы AI-кодинга без TDD
+
+1. **Hallucinations и incorrect implementations** — AI генерирует код, который компилируется но не работает правильно
+2. **Scope creep** — AI добавляет "полезные" фичи, которые не требовались
+3. **Context rot** — AI теряет контекст в длинных сессиях, повторяет ошибки
+4. **Untestable code** — AI генерирует tightly coupled код без учёта тестируемости
+5. **Accumulating technical debt** — "move fast and break things" без safety net
+6. **Developer skill atrophy** — разработчики становятся copy-paste operators
+7. **Ambiguous requirements** — AI интерпретирует prompt по-своему
+
+### Решение через TDD
+
+- **Binary validation** → hallucinations caught immediately
+- **Tests define scope** → AI cannot exceed test requirements
+- **Tests = persistent context** → no context rot, reusable across sessions
+- **Design for testability** → forces better architecture
+- **Regression prevention** → existing tests catch breaks
+- **Developer stays architect** → designs behavior, AI implements
+- **Unambiguous spec** → failing test = clear goal
+
+---
+
+## Авторитетные источники
+
+### Thoughtworks Technology Radar (2023-2025)
+
+- **[AI-aided test-first development](https://www.thoughtworks.com/en-us/radar/techniques/ai-aided-test-first-development)** (апрель 2023)
+- **Статус**: TRIAL — "we've seen promising results"
+- **Применение**: junior developers + AI align с team's architectural style
+- **Подход**: generate tests with ChatGPT → developer implements → или AI implements с test-first constraint
+
+### DORA Report 2025 (Google Cloud)
+
+- **[State of AI-Assisted Software Development 2025](https://dora.dev/research/2025/dora-report/)**
+- **Главная находка**: "AI is an amplifier, not a fix" — усиливает существующие практики
+- **TDD impact**: организации с strong testing practices получают больше выгоды от AI
+- **[Google Cloud Blog](https://cloud.google.com/discover/how-test-driven-development-amplifies-ai-success)** (январь 2026): "TDD amplifies AI success"
+- **Цитата**: "Quality practices such as TDD help organizations harness AI's productivity gains while maintaining stability"
+
+### Kent Beck (создатель TDD и XP)
+
+- **[Pragmatic Engineer Podcast](https://newsletter.pragmaticengineer.com/p/tdd-ai-agents-and-coding-with-kent)** (июнь 2025)
+- **[YouTube Interview](https://www.youtube.com/watch?v=aSXaxOdVtAQ)** "TDD, AI agents and coding with Kent Beck"
+- **Цитата**: "TDD is a superpower when working with AI. I communicate things the Genie missed in terms of tests"
+- **Практика**: запускает тесты каждые 300ms чтобы "catch the genie"
+- **LinkedIn**: "The genie doesn't want to do TDD. It wants to go fast. You have to force red-green-refactor"
+
+### 8th Light (Software Consultancy)
+
+- **[TDD: The Missing Protocol for Effective AI Collaboration](https://8thlight.com/insights/tdd-effective-ai-collaboration)** (июль 2025)
+- **Тезис**: TDD breaks problems into small, testable behaviors → structured context для AI
+- **Вывод**: TDD не просто best practice, это communication protocol
+
+### Builder.io Engineering
+
+- **[Test-Driven Development with AI](https://www.builder.io/blog/test-driven-development-ai)** (август 2025)
+- **Тезис**: "AI turns TDD's weaknesses into strengths"
+- **Анализ**: AI thrives on clear, measurable goals — binary test = clearest goal
+- **Практика**: AI generates boilerplate tests, developer focuses на edge cases
+
+---
+
+## Формальные спеки и фреймворки
+
+### AgileМania: AI-Enabled TDD Certification
+
+- **[AI-Enabled Test-Driven Development (AI-TDD) Training](https://agilemania.com/test-driven-development-tdd-with-ai-training)** (декабрь 2024)
+- **Формат**: 1-day hands-on certification course
+- **Охват**: fundamentals of TDD + AI pair programming + hands-on exercises
+- **Статистика**: "By 2025, 85% of software projects incorporate AI tools"
+- **Особенности**: фокус на практическую интеграцию GitHub Copilot и Claude в TDD workflow
+
+### Galileo AI: TDD for AI Systems Framework
+
+- **[Test-Driven Development for AI Systems](https://galileo.ai/blog/test-driven-development-ai-systems)** (январь 2026)
+- **[TDD for AI System Architecture](https://galileo.ai/blog/tdd-ai-system-architecture)** (август 2025)
+- **Особенности**: framework для тестирования самих AI-систем, не AI-assisted coding
+- **Компоненты**:
+  - Model Quality Checklist
+  - Test-First Specification Template
+  - AI Component Testing Matrix
+- **Применение**: production AI systems с guardrails
+
+### Thoughtworks Structured Approach
+
+- **[Technology Radar Entry](https://www.thoughtworks.com/en-us/radar/techniques/ai-aided-test-first-development)** (2023)
+- **Методология**:
+  1. Reusable prompt "fragments" (tech stack + design patterns + team conventions)
+  2. Feature description + acceptance criteria
+  3. AI generates implementation plan → tests → code
+- **Особенности**: enterprise-scale применение, интеграция в CI/CD
+
+### Nimble Approach: TDD Protocol for AI
+
+- **[How to Use TDD for Better AI Coding Outputs](https://nimbleapproach.com/blog/how-to-use-test-driven-development-for-better-ai-coding-outputs/)** (ноябрь 2025)
+- **Особенности**: пошаговый protocol для AI prompting с TDD
+- **Инструкции**: как структурировать prompts для генерации тестов первыми
+
+### Qodo (ex-CodiumAI): AI Test Generation
+
+- **[AI Code Assistants Revolutionizing TDD](https://www.qodo.ai/blog/ai-code-assistants-test-driven-development/)** (июль 2025)
+- **Продукт**: Qodo Gen — AI test generation tool
+- **Особенности**: интегрируется в IDE, генерирует unit tests automatically
+- **Поддержка**: Python, JavaScript, TypeScript, Java
+
+---
+
+## Использование в Claude Code
+
+### Нативная поддержка
+
+**Claude Code** полностью поддерживает AI-aided test-first подход через:
+
+1. **Explicit test-first prompts**:
+   ```
+   "Создай failing test для функции calculateTax(), которая должна 
+   обрабатывать случаи: positive income, zero income, negative income"
+   ```
+
+2. **Output styles для TDD**:
+   - `/output-style learning` — Claude объясняет почему тест failing и как его починить
+   - `/output-style explanatory` — подробное объяснение test design decisions
+
+3. **MCP integration** — подключение test runners через Model Context Protocol
+
+### Workflow в Claude Code
+
+**Step 1: Test Requirements**
+```
+Prompt: "Напиши тесты для user authentication module:
+- Valid credentials → success
+- Invalid password → error with message
+- Nonexistent user → error with message
+- Empty fields → validation error"
+```
+
+**Step 2: Claude генерирует failing tests**
+```typescript
+describe('AuthService', () => {
+  test('authenticates user with valid credentials', async () => {
+    const result = await authService.login('user@example.com', 'password123')
+    expect(result.success).toBe(true)
+    expect(result.user).toBeDefined()
+  })
+  // ... остальные тесты
+})
+```
+
+**Step 3: Запуск (все failing)**
+```
+Prompt: "Запусти тесты и покажи результаты"
+Claude: *executes tests, shows all failures*
+```
+
+**Step 4: Implementation**
+```
+Prompt: "Имплементируй AuthService чтобы пройти эти тесты"
+Claude: *generates minimal implementation*
+```
+
+**Step 5: Validation**
+```
+Prompt: "Запусти тесты снова"
+Claude: *all tests pass*
+```
+
+**Step 6: Refactor**
+```
+Prompt: "Рефактор: выдели error handling в отдельный класс"
+Claude: *refactors with tests as safety net*
+```
+
+### Best Practices для Claude Code
+
+- **Используй Projects** — добавь testing framework docs в project knowledge
+- **Custom instructions**: "Always generate tests before implementation. Run tests after each change."
+- **Iterative approach** — один тест → implementation → следующий тест
+- **Explicit commands**: "Run tests", "Show test coverage", "Refactor without breaking tests"
+
+### Источники
+
+- [Anthropic Engineering: Claude Code Best Practices](https://www.anthropic.com/engineering/claude-code-best-practices) (апрель 2025)
+- [Pair Programming with Claude Code](https://shipyard.build/blog/claude-code-output-styles-pair-programming/) (сентябрь 2025)
+
+---
+
+## Использование в Cursor
+
+### Нативная поддержка
+
+**Cursor** имеет специальные возможности для TDD workflow:
+
+1. **Composer Agent Mode** — multi-step task execution с test validation
+2. **Inline test generation** — `Cmd/Ctrl + K` → "generate test for this function"
+3. **Auto-run tests** — Cursor автоматически запускает тесты после генерации кода
+4. **Test-aware refactoring** — проверяет, что рефакторинг не ломает тесты
+
+### Workflow в Cursor
+
+**Method 1: Inline TDD**
+
+1. Напиши comment с test requirements: `// Test: calculateDiscount should return 10% for orders > $100`
+2. `Cmd + K` → "Generate test from this comment"
+3. Cursor генерирует failing test
+4. `Cmd + K` на функции → "Implement to pass the test above"
+5. Cursor автоматически запускает тесты
+
+**Method 2: Composer TDD**
+
+1. Открой Composer (`Cmd + I`)
+2. Prompt: "Create feature X using TDD: write tests first, then implement"
+3. Composer генерирует test file → implementation file → runs tests → показывает результаты
+4. Если тесты не проходят, Composer автоматически исправляет
+
+**Method 3: Chat TDD**
+
+1. Chat panel: "I want to add feature Y. Let's start with tests."
+2. Cursor генерирует test scenarios
+3. Ты выбираешь какие тесты написать first
+4. Cursor генерирует failing tests
+5. "Now implement to pass these tests"
+6. Cursor имплементирует + auto-runs tests
+
+### Cursor-специфичные фичи
+
+- **Multi-file test generation** — Cursor одновременно создаёт test file + implementation stub
+- **Test coverage view** — показывает какие части кода покрыты тестами
+- **Smart refactoring** — предлагает рефакторинг только если тесты проходят
+- **Voice mode TDD** — можно проговаривать test requirements голосом
+
+### Best Practices для Cursor
+
+- **Rules for AI** (`.cursorrules`):
+  ```
+  Always use test-driven development:
+  1. Generate failing tests first
+  2. Implement minimal code to pass
+  3. Refactor with tests as safety net
+  4. Run tests after every change
+  ```
+- **Composer для complex features** — пусть agent управляет TDD циклом
+- **Inline для quick fixes** — fast test generation для single functions
+- **Git worktrees** — Cursor может запускать TDD workflow в изолированных branches
+
+### Источники
+
+- [Cursor Official Docs](https://cursor.com)
+- [Launchpad Lab: AI Coding with Windsurf - A New Approach to TDD](https://launchpadlab.com/blog/ai-coding-with-windsurf-a-new-approach-to-tdd/) (июнь 2025) — применимо к Cursor
+- [Skywork AI: Cursor's AI Autocomplete in Pair Programming](https://skywork.ai/blog/cursor-1-7-ai-autocomplete-pair-programming-best-practices/) (сентябрь 2025)
+
+---
+
+## Ссылки
+
+### Официальные источники
+
+- [Thoughtworks Technology Radar: AI-aided test-first development](https://www.thoughtworks.com/en-us/radar/techniques/ai-aided-test-first-development) (апрель 2023)
+- [DORA Report 2025: State of AI-Assisted Software Development](https://dora.dev/research/2025/dora-report/)
+- [Google Cloud: How TDD Amplifies AI Success](https://cloud.google.com/discover/how-test-driven-development-amplifies-ai-success) (январь 2026)
+
+### Авторитетные голоса
+
+- [Kent Beck: TDD, AI agents and coding (Podcast)](https://newsletter.pragmaticengineer.com/p/tdd-ai-agents-and-coding-with-kent) (июнь 2025)
+- [Kent Beck YouTube Interview](https://www.youtube.com/watch?v=aSXaxOdVtAQ)
+- [8th Light: TDD as Missing Protocol for AI](https://8thlight.com/insights/tdd-effective-ai-collaboration) (июль 2025)
+- [Builder.io: Test-Driven Development with AI](https://www.builder.io/blog/test-driven-development-ai) (август 2025)
+
+### Спеки и фреймворки
+
+- [AgileМania: AI-Enabled TDD Training](https://agilemania.com/test-driven-development-tdd-with-ai-training)
+- [Galileo AI: TDD for AI Systems](https://galileo.ai/blog/test-driven-development-ai-systems) (январь 2026)
+- [Nimble Approach: TDD Protocol for AI](https://nimbleapproach.com/blog/how-to-use-test-driven-development-for-better-ai-coding-outputs/) (ноябрь 2025)
+- [Qodo: AI Code Assistants & TDD](https://www.qodo.ai/blog/ai-code-assistants-test-driven-development/) (июль 2025)
+
+### Практические руководства
+
+- [Anthropic: Claude Code Best Practices](https://www.anthropic.com/engineering/claude-code-best-practices) (апрель 2025)
+- [Atomic Spin: How I Made AI Respect TDD](https://spin.atomicobject.com/make-ai-respect-tdd/) (январь 2026)
+- [Ready Set Cloud: TDD with AI - The Right Way](https://www.readysetcloud.io/blog/allen.helton/tdd-with-ai/) (октябрь 2023)
+
+### Дополнительные ресурсы
+
+- [Phil Parsons: Relevance of TDD in AI Workflow](https://philparsons.co.uk/blog/the-relevance-of-tdd-in-an-ai-workflow/)
+- [Code Manship: Why TDD Works in AI Programming](https://codemanship.wordpress.com/2026/01/09/why-does-test-driven-development-work-so-well-in-ai-assisted-programming/) (январь 2026)
+- [NOPC Accelerate: TDD Guide 2025](https://www.nopaccelerate.com/test-driven-development-guide-2025/) (сентябрь 2025)
diff --git a/assets/beyond-vibe-coding/images/comic/3a453d3b-6755-4b78-9151-5fa9d2dde9c1.jpeg b/assets/beyond-vibe-coding/images/comic/3a453d3b-6755-4b78-9151-5fa9d2dde9c1.jpeg
new file mode 100644
index 0000000..93ce120
Binary files /dev/null and b/assets/beyond-vibe-coding/images/comic/3a453d3b-6755-4b78-9151-5fa9d2dde9c1.jpeg differ
diff --git a/assets/beyond-vibe-coding/images/comic/Generated Image February 01, 2026 - 6_20PM.jpeg b/assets/beyond-vibe-coding/images/comic/Generated Image February 01, 2026 - 6_20PM.jpeg
new file mode 100644
index 0000000..fd8025f
Binary files /dev/null and b/assets/beyond-vibe-coding/images/comic/Generated Image February 01, 2026 - 6_20PM.jpeg differ
diff --git a/assets/beyond-vibe-coding/images/comic/Generated Image February 01, 2026 - 6_48PM.jpeg b/assets/beyond-vibe-coding/images/comic/Generated Image February 01, 2026 - 6_48PM.jpeg
new file mode 100644
index 0000000..75d68b3
Binary files /dev/null and b/assets/beyond-vibe-coding/images/comic/Generated Image February 01, 2026 - 6_48PM.jpeg differ
diff --git a/assets/beyond-vibe-coding/images/comic/Generated Image February 01, 2026 - 7_10PM.jpeg b/assets/beyond-vibe-coding/images/comic/Generated Image February 01, 2026 - 7_10PM.jpeg
new file mode 100644
index 0000000..600798d
Binary files /dev/null and b/assets/beyond-vibe-coding/images/comic/Generated Image February 01, 2026 - 7_10PM.jpeg differ
diff --git a/assets/beyond-vibe-coding/images/comic/Generated Image January 31, 2026 - 10_53PM.jpeg b/assets/beyond-vibe-coding/images/comic/Generated Image January 31, 2026 - 10_53PM.jpeg
new file mode 100644
index 0000000..3e65ccf
Binary files /dev/null and b/assets/beyond-vibe-coding/images/comic/Generated Image January 31, 2026 - 10_53PM.jpeg differ
diff --git a/assets/beyond-vibe-coding/images/comic/reg/1.jpeg.jpg b/assets/beyond-vibe-coding/images/comic/reg/1.jpeg.jpg
new file mode 100644
index 0000000..f28d1c8
Binary files /dev/null and b/assets/beyond-vibe-coding/images/comic/reg/1.jpeg.jpg differ
diff --git a/assets/beyond-vibe-coding/images/comic/reg/2.jpg b/assets/beyond-vibe-coding/images/comic/reg/2.jpg
new file mode 100644
index 0000000..b8f3034
Binary files /dev/null and b/assets/beyond-vibe-coding/images/comic/reg/2.jpg differ
diff --git a/assets/beyond-vibe-coding/images/comic/reg/Generated Image January 31, 2026 - 11_18PM.jpeg b/assets/beyond-vibe-coding/images/comic/reg/Generated Image January 31, 2026 - 11_18PM.jpeg
new file mode 100644
index 0000000..6ce3836
Binary files /dev/null and b/assets/beyond-vibe-coding/images/comic/reg/Generated Image January 31, 2026 - 11_18PM.jpeg differ
diff --git a/assets/beyond-vibe-coding/images/infographic/Generated Image January 31, 2026 - 11_11PM.jpeg b/assets/beyond-vibe-coding/images/infographic/Generated Image January 31, 2026 - 11_11PM.jpeg
new file mode 100644
index 0000000..f5b962a
Binary files /dev/null and b/assets/beyond-vibe-coding/images/infographic/Generated Image January 31, 2026 - 11_11PM.jpeg differ
diff --git a/assets/beyond-vibe-coding/log-chat.md b/assets/beyond-vibe-coding/log-chat.md
index c319187..f4bac46 100644
--- a/assets/beyond-vibe-coding/log-chat.md
+++ b/assets/beyond-vibe-coding/log-chat.md
@@ -384,7 +384,7 @@ After validation PASS → @writer creates Draft using corrected structure
 
 ---
 
-@user
+## 2026-01-31 @user
 
 я добавил файл со статистикой по применению AI в коде. позже нам нужно будет использовать его для создания инфорграфики. В текст статьи я вставил TODOs с детальным описанием.
 Файл называется [ai-adoption-statistics](ai-adoption-statistics.md)
@@ -472,4 +472,88 @@ After validation PASS → @writer creates Draft using corrected structure
 
 ---
 
+## 2026-01-31 @seo — TDD Section Keyword Research
+
+**Action:** Keyword research for TDD methodology naming
+
+**Problem:** Official name "AI-aided test-first development" (Thoughtworks) has **zero** search volume. People don't search for it.
+
+**Research completed:**
+- DataForSEO keyword research: ~$0.15 spent
+- Tested 30+ variations of TDD + AI naming
+
+**Key findings:**
+
+| Keyword | Volume | Comp | Notes |
+|---------|--------|------|-------|
+| tdd | 6,600 | LOW | 🔥 Parent keyword |
+| test driven development | 3,600 | LOW | Core term, KD 41 |
+| red green refactor | 260 | LOW | TDD cycle, Kent Beck |
+| test driven development python | 210 | LOW | Language-specific |
+| test first development | 70 | LOW | Synonym |
+| tdd example | 70 | LOW | Tutorial intent |
+| kent beck tdd | 50 | LOW | Authority signal |
+| **test driven development with ai** | 40 | HIGH | 🎯 Exact match! |
+| tdd vibe coding | 10 | MED | 🚀 Emerging! |
+| vibe coding tdd | 10 | LOW | 🚀 Emerging! |
+| ai-aided test-first development | 0 | — | ❌ Official but dead |
+| ai-assisted tdd | 0 | — | ❌ |
+| ai-enabled tdd | 0 | — | ❌ |
+
+**Naming analysis:**
+
+| Name | Source | Volume | Verdict |
+|------|--------|--------|--------|
+| AI-aided test-first development | Thoughtworks | 0 | ❌ SEO-dead |
+| Test-Driven Development with AI | Builder.io | 40 | ✅ Best match |
+| AI-assisted TDD | Industry | 0 | ❌ |
+| TDD + AI | Shorthand | — | ✅ Leverages parent |
+
+**Decision:**
+
+**H2:** `TDD + AI: Quality-First Approach`
+
+Rationale:
+- "TDD" = 6,600 vol parent keyword (LOW competition)
+- "+AI" gives context
+- "Quality-First" contrasts with vibe coding
+- Short, scannable
+
+**First paragraph template:**
+
+> Test-driven development (TDD) — the red-green-refactor cycle that Kent Beck popularized — becomes a superpower when combined with AI coding. Unlike vibe coding where you trust the output, TDD makes tests your communication protocol with AI. Test first development forces clear specification before implementation...
+
+**Keywords covered in first 100 words:**
+- ✅ test-driven development (3,600)
+- ✅ TDD (6,600)
+- ✅ red-green-refactor (260)
+- ✅ Kent Beck (50)
+- ✅ vibe coding (90,500) — contrast
+- ✅ test first development (70)
+
+**Note:** "Quality-First Approach" has only 10 vol, but all alternatives ("test first approach", "quality first development") are 0-10 vol too. Keep it — good messaging that contrasts with vibe coding.
+
+**Emerging opportunity:**
+"tdd vibe coding" and "vibe coding tdd" — 10 vol each but growing. People starting to search for TDD + vibe coding connection. Include both terms in text for future ranking.
+
+**For @writer:**
+- Update TDD section H2 to `TDD + AI: Quality-First Approach`
+- Rewrite first paragraph using template above
+- Mention "vibe coding tdd" contrast organically
+
+---
+
+## 2026-01-31 @user - potential additions
+
+**For @writer:**
+
+1 Добавь еще TODO в индексный файл рассмотреть еще одну часть со следующими разделами:
+
+- Architecture-First AI Development
+- Prompt-Driven Development
+- Copy-pasting from AI chatbot
+
+2 Посмотри все md файлы в этой папке. Там добавились спеки по конкретным методологиям. добавь их в индекс и исопльзуй для расширения раздела credentials чтобы больше деталей дать. вставляй в текст статьи ссылки на пруфы и спеки компаний которые продвигают соотвествующие подходы.
+
+---
 
diff --git a/assets/beyond-vibe-coding/text.md b/assets/beyond-vibe-coding/text.md
index 637a22d..663ebdb 100644
--- a/assets/beyond-vibe-coding/text.md
+++ b/assets/beyond-vibe-coding/text.md
@@ -30,7 +30,7 @@ Let's figure out what exactly we can do with AI. Different approaches exist, giv
 - No upfront planning, minimal specification
 - Trust AI to handle details, fix issues as they appear
 
-Vibe coding is a great approach. Really. I use it often myself. It works perfectly for non-critical features, dev tools, prototypes, experiments.
+Vibe coding is a great approach. Really. I use it often myself. It works perfectly for non-critical features, dev tools, prototypes, experiments. [правда я все же использую клавиатуру для этого]
 
 When do I use it?
 - When the result is easy to evaluate visually