wizz-method 2.1.1 → 2.3.0

package/README.md

@@ -4,120 +4,120 @@
   <a href="https://www.npmjs.com/package/wizz-method"><img alt="npm" src="https://img.shields.io/npm/v/wizz-method?color=FF4500&label=wizz-method&labelColor=0D0D0D"></a>
   <a href="https://nodejs.org"><img alt="Node.js" src="https://img.shields.io/badge/node-%3E%3D20.12-FF4500?labelColor=0D0D0D"></a>
   <a href="https://method.wizzcomms.com"><img alt="Docs" src="https://img.shields.io/badge/docs-method.wizzcomms.com-FF4500?labelColor=0D0D0D"></a>
-  <a href="https://method.wizzcomms.com"><img alt="Licença" src="https://img.shields.io/badge/licen%C3%A7a-propriet%C3%A1ria-0D0D0D?labelColor=FF4500"></a>
+  <a href="https://method.wizzcomms.com"><img alt="License" src="https://img.shields.io/badge/license-proprietary-0D0D0D?labelColor=FF4500"></a>
 </p>
 
 # Wizz Method
 
-**O método de agência orientado por IA, em português.** Você descreve o pedido em linguagem normal. Um router lê, descobre a área e chama o agente certo. Quando o trabalho cruza áreas, o Maestro monta a sequência e cobra o fechamento. Tudo dentro do assistente de IA que você já usa.
+**The AI-driven agency method, built for Brazilian Portuguese.** You describe the request in plain language. A router reads it, figures out the area, and calls the right agent. When the work crosses areas, the Maestro sequences it and drives it to completion. All inside the AI assistant you already use.
 
-Cansado de reexplicar o contexto a cada conversa? O Wizz Method dá ao seu time uma estrutura fixa: agentes especializados por área, skills curadas por trás de cada agente, memória persistente do projeto e um encerramento padrão em toda resposta (**✅ o que fiz**, **➡️ próximo passo**, **🎯 comando**).
+Tired of re-explaining context in every conversation? Wizz Method gives your team a fixed structure: specialized agents per area, curated skills behind each agent, persistent project memory, and a standard closing on every response (**✅ what I did**, **➡️ next step**, **🎯 command**). The agents work in, and respond to you in, Brazilian Portuguese by default.
 
-## Como funciona
+## How it works
 
 ```mermaid
 flowchart LR
-    P[Seu pedido] --> R[wizz-router<br/>triagem]
-    R -->|1 área, pontual| A[Agente da área]
-    R -->|2+ áreas ou multi-passo| M[wizz-maestro<br/>orquestra]
+    P[Your request] --> R[wizz-router<br/>triage]
+    R -->|1 area, one-off| A[Area agent]
+    R -->|2+ areas or multi-step| M[wizz-maestro<br/>orchestrates]
     M --> A
-    A --> S[Skills, CLIs e MCPs<br/>do catálogo]
-    S --> F[Entrega + memória<br/>✅ ➡️ 🎯]
+    A --> S[Skills, CLIs and MCPs<br/>from the catalog]
+    S --> F[Delivery + memory<br/>✅ ➡️ 🎯]
 ```
 
-- **`wizz-router`** é a porta de entrada. Ele não executa: mede a complexidade e delega.
-- **`wizz-maestro`** entra quando o pedido cruza áreas ou precisa de planejamento. Coordena os agentes e garante que a decisão vai para a memória.
-- **Cada agente** conhece só a própria área e puxa as skills, CLIs e MCPs registrados para ela no catálogo. Nada é improvisado.
+- **`wizz-router`** is the entry point. It doesn't execute: it measures complexity and delegates.
+- **`wizz-maestro`** steps in when the request crosses areas or needs planning. It coordinates the agents and makes sure the decision reaches memory.
+- **Each agent** knows only its own area and pulls the skills, CLIs and MCPs registered for it in the catalog. Nothing is improvised.
 
-## Os agentes
+## The agents
 
-| Área | Agente | Cuida de |
+| Area | Agent | Handles |
 |---|---|---|
-| 🧭 Orquestração | `wizz-maestro` | Lê o pedido e despacha o agente certo |
-| 🎨 Design | `wizz-designer` | Visual, landing pages, motion, 3D, componentes |
-| ✍️ Copy | `wizz-copy` | Copy, e-mail, headlines, humanização de texto |
-| 🔍 SEO | `wizz-seo` | SEO tradicional e de IA, schema, arquitetura de site |
-| 📈 Growth | `wizz-growth` | Marketing, CRO, lançamento, preço, churn |
-| 📣 Social | `wizz-social` | Conteúdo e distribuição em redes |
-| 📢 Ads | `wizz-ads` | Mídia paga, criativos, tracking |
-| ✅ QA | `wizz-qa` | Revisão adversarial, testes E2E, qualidade da entrega |
-| 🧠 Memória | `wizz-memoria` | Cérebro do projeto: decisões, estado, sessões |
+| 🧭 Orchestration | `wizz-maestro` | Reads the request and dispatches the right agent |
+| 🎨 Design | `wizz-designer` | Visual, landing pages, motion, 3D, components |
+| ✍️ Copy | `wizz-copy` | Copy, email, headlines, text humanization |
+| 🔍 SEO | `wizz-seo` | Traditional and AI SEO, schema, site architecture |
+| 📈 Growth | `wizz-growth` | Marketing, CRO, launches, pricing, churn |
+| 📣 Social | `wizz-social` | Social content and distribution |
+| 📢 Ads | `wizz-ads` | Paid media, creatives, tracking |
+| ✅ QA | `wizz-qa` | Adversarial review, E2E tests, delivery quality |
+| 🧠 Memory | `wizz-memoria` | Project cerebro: decisions, state, sessions |
 
-Os papéis de produto e engenharia (analista, PM, arquiteto, dev, QA, UX, tech writer) são agentes do módulo `bmm`, personalizados em PT-BR pelos overrides em `src/modules/wizz/overrides/`. O catálogo completo, com as skills de cada área, está em `skills-registry.yaml`.
+Product and engineering roles (analyst, PM, architect, dev, QA, UX, tech writer) are agents from the `bmm` module, customized to speak Brazilian Portuguese through the overrides in `src/modules/wizz/overrides/`. The full catalog, with each area's skills, lives in `skills-registry.yaml`.
 
-A camada Wizz vale para todos: encerramento padrão (✅/➡️/🎯), economia de token (Cérebro primeiro, depois grep, só então abrir arquivo), memória carregada na ativação e idioma à sua escolha. Detalhes no README do módulo, em `src/modules/wizz/README.md`.
+The Wizz layer applies to everyone: standard closing (✅/➡️/🎯), token economy (cerebro first, then grep, only then opening the file), memory loaded on activation, and your choice of language. Details in the module's README, at `src/modules/wizz/README.md`.
 
-## O que vem no pacote
+## What's in the package
 
-| Camada | O que é | Como se prova |
+| Layer | What it is | How it's proven |
 |---|---|---|
-| Catálogo | `skills-registry.yaml` como fonte única de áreas, agentes, skills, CLIs e MCPs | `npm run validate:registry-schema` |
-| Agentes resolvidos | Cada agente já chega pré-resolvido em `_wizz/_resolved/`, sem Python na ativação | `test/test-resolved-customizations.js` |
-| Hooks | Roteamento obrigatório, regras de sessão, trace de decisão e contexto de segurança, registrados pelo installer no `.claude/settings.json` do projeto | `npx wizz-method doctor` |
-| Installer | Instalação e atualização atômicas, módulos externos pinados por tag e SHA | `test/test-pin-sha.js` |
-| Memória | `memlog.js` sem dependências, Cérebro por projeto, protocolo de handoff entre agentes | `test/test-memlog.js` |
-| Diagnóstico | `wizz doctor` checa hooks, rtk, jq, Python (opcional) e drift em `_wizz/` | `npx wizz-method doctor` |
+| Catalog | `skills-registry.yaml` as the single source of areas, agents, skills, CLIs and MCPs | `npm run validate:registry-schema` |
+| Resolved agents | Every agent arrives pre-resolved in `_wizz/_resolved/`, no Python at activation | `test/test-resolved-customizations.js` |
+| Hooks | Mandatory routing, session rules, decision trace and security context, registered by the installer in the project's `.claude/settings.json` | `npx wizz-method doctor` |
+| Installer | Atomic install and update, external modules pinned by tag and SHA | `test/test-pin-sha.js` |
+| Memory | Dependency-free `memlog.js`, per-project cerebro, handoff protocol between agents | `test/test-memlog.js` |
+| Diagnostics | `wizz doctor` checks hooks, rtk, jq, Python (optional) and drift in `_wizz/` | `npx wizz-method doctor` |
 
 ## Quick Start
 
-**Pré-requisito único:** [Node.js](https://nodejs.org) 20.12 ou superior. Instalar, atualizar e ativar qualquer agente roda só com Node.
+**Single prerequisite:** [Node.js](https://nodejs.org) 20.12 or later. Installing, updating and activating any agent runs on Node alone.
 
-[Python](https://www.python.org) 3.11+ e [uv](https://docs.astral.sh/uv/) são opcionais e servem só às skills que trazem script próprio (`wizz-brainstorming`, `wizz-retrospective`, `wizz-sprint-planning`, `wizz-architecture`, `wizz-party-mode`, `wizz-forge-idea`, `wizz-advanced-elicitation`, `wizz-customize`). Sem eles o resto funciona igual, e o `doctor` lista o check de Python como opcional.
+[Python](https://www.python.org) 3.11+ and [uv](https://docs.astral.sh/uv/) are optional and only serve skills that ship their own script (`wizz-brainstorming`, `wizz-retrospective`, `wizz-sprint-planning`, `wizz-architecture`, `wizz-party-mode`, `wizz-forge-idea`, `wizz-advanced-elicitation`, `wizz-customize`). Everything else works the same without them, and `doctor` lists the Python check as optional.
 
-**Plataformas:** macOS, Linux e WSL. Windows nativo (cmd.exe ou PowerShell sem WSL) não é suportado.
+**Platforms:** macOS, Linux and WSL. Native Windows (cmd.exe or PowerShell without WSL) is not supported.
 
-### 1. Instale no projeto
+### 1. Install in your project
 
 ```bash
 npx wizz-method install
 ```
 
-Siga os prompts, depois abra seu IDE de IA (Claude Code, Cursor, Codex e outros) na pasta do projeto.
+Follow the prompts, then open your AI IDE (Claude Code, Cursor, Codex and others) in the project folder.
 
-### 2. Personalize a camada Wizz
+### 2. Customize the Wizz layer
 
-Pergunta o idioma dos agentes e aplica os overrides PT-BR. É idempotente.
+Asks for the agents' language and applies the Brazilian Portuguese overrides. It's idempotent.
 
 ```bash
-node <caminho-do-wizz-method>/src/modules/wizz/scripts/wizz-init.mjs .
+node <path-to-wizz-method>/src/modules/wizz/scripts/wizz-init.mjs .
 ```
 
-### 3. Confira a instalação
+### 3. Check the install
 
 ```bash
 npx wizz-method doctor
 ```
 
-### 4. Use
+### 4. Use it
 
-Invoque o `wizz-router` e mande seu pedido. Para trabalhos que cruzam áreas, chamar o `wizz-maestro` direto também funciona, mas o router é a entrada canônica.
+Invoke `wizz-router` and send your request. For work that crosses areas, calling `wizz-maestro` directly also works, but the router is the canonical entry point.
 
-### Instalação não interativa (CI/CD)
+### Non-interactive install (CI/CD)
 
 ```bash
-npx wizz-method install --directory /caminho/do/projeto --modules bmm --tools claude-code --yes
-node <caminho-do-wizz-method>/src/modules/wizz/scripts/wizz-init.mjs . --lang "English"
+npx wizz-method install --directory /path/to/project --modules bmm --tools claude-code --yes
+node <path-to-wizz-method>/src/modules/wizz/scripts/wizz-init.mjs . --lang "English"
 ```
 
-Sem TTY, o idioma padrão é "Português (BR)". Use `--lang` ou a variável `WIZZ_LANG` para definir outro.
+Without a TTY, the default language is "Português (BR)". Use `--lang` or the `WIZZ_LANG` environment variable to set another one.
 
-## Comandos
+## Commands
 
-| Comando | Faz |
+| Command | Does |
 |---|---|
-| `npx wizz-method install` | Instala módulos, hooks, skills e MCPs no projeto |
-| `npx wizz-method status` | Mostra o que está instalado e a versão de cada módulo |
-| `npx wizz-method doctor` | Diagnóstico: hooks, rtk, jq, Python opcional e drift de `_wizz/` |
-| `npx wizz-method resolve [--skill <nome>]` | Regenera os snapshots de customização em `_wizz/_resolved/` |
-| `npx wizz-method fetch-assets` | Baixa sob demanda os assets pesados (áudio e fontes) |
-| `npx wizz-method uninstall` | Remove a instalação do projeto |
+| `npx wizz-method install` | Installs modules, hooks, skills and MCPs into the project |
+| `npx wizz-method status` | Shows what's installed and each module's version |
+| `npx wizz-method doctor` | Diagnostics: hooks, rtk, jq, optional Python and `_wizz/` drift |
+| `npx wizz-method resolve [--skill <name>]` | Regenerates the customization snapshots in `_wizz/_resolved/` |
+| `npx wizz-method fetch-assets` | Downloads heavy assets on demand (audio and fonts) |
+| `npx wizz-method uninstall` | Removes the install from the project |
 
-## Documentação e suporte
+## Documentation and support
 
-- Docs: [method.wizzcomms.com](https://method.wizzcomms.com), com tutoriais, guias, conceitos e referência.
-- Suporte: <support@wizzcomms.com>.
-- Vulnerabilidades: veja o arquivo `SECURITY.md` do pacote.
+- Docs: [method.wizzcomms.com](https://method.wizzcomms.com), with tutorials, guides, concepts and reference.
+- Support: <support@wizzcomms.com>.
+- Vulnerabilities: see the package's `SECURITY.md` file.
 
-## Licença
+## License
 
-O Wizz Method é software proprietário da **Wizz! comms.**, licenciado mediante contrato comercial. Veja o arquivo `LICENSE`. Os componentes de terceiros e seus avisos de licença estão em `THIRD-PARTY-NOTICES.md`, no mesmo pacote.
+Wizz Method is proprietary software owned by **Wizz! comms.**, licensed under a commercial agreement. See the `LICENSE` file. Third-party components and their license notices are in `THIRD-PARTY-NOTICES.md`, in the same package.

package/SECURITY.md

@@ -1,24 +1,23 @@
-# Política de Segurança
+# Security Policy
 
-## Como reportar uma vulnerabilidade
+## How to report a vulnerability
 
-Encontrou uma vulnerabilidade de segurança no Wizz Method? Não abra uma
-issue pública. Reporte por e-mail para <support@wizzcomms.com>, com o
-máximo de detalhe possível (tipo de vulnerabilidade, arquivos envolvidos,
-passos para reproduzir e impacto estimado).
+Found a security vulnerability in Wizz Method? Do not open a public
+issue. Report it by email to <support@wizzcomms.com>, with as much
+detail as possible (type of vulnerability, files involved, steps to
+reproduce, and estimated impact).
 
-## Prazo de resposta
+## Response time
 
-Respondemos em até **48 horas úteis** após o recebimento do relato e damos
-uma atualização de status em até 7 dias.
+We respond within **48 business hours** of receiving the report and give
+a status update within 7 days.
 
-## Versões suportadas
+## Supported versions
 
-Mantemos patches de segurança apenas para a **última versão publicada**
-(`latest` no npm).
+We maintain security patches only for the **latest published version**
+(`latest` on npm).
 
-## Divulgação responsável
+## Responsible disclosure
 
-Pedimos que não seja feita divulgação pública da vulnerabilidade antes que
-uma correção esteja disponível. Agradecemos a colaboração de quem reporta em
-manter o Wizz Method seguro.
+We ask that the vulnerability not be publicly disclosed before a fix is
+available. We appreciate reporters' help in keeping Wizz Method secure.

package/THIRD-PARTY-NOTICES.md

@@ -1,20 +1,20 @@
-# Avisos de terceiros
+# Third-party notices
 
-O Wizz Method incorpora código e conteúdo de terceiros, listados abaixo com
-sua origem, licença e localização no repositório. Os termos de cada licença
-listada aqui prevalecem sobre a [LICENSE](LICENSE) para o respectivo
-componente. Dependências de `package.json` (node_modules) não entram neste
-inventário: seguem sob suas próprias licenças, resolvidas pelo npm.
+Wizz Method incorporates third-party code and content, listed below with
+its origin, license, and location in the repository. The terms of each
+license listed here take precedence over the [LICENSE](LICENSE) for the
+respective component. Dependencies from `package.json` (node_modules) are
+not part of this inventory: they follow their own licenses, resolved by npm.
 
 ## BMAD Method (BMad Code, LLC)
 
-O motor de workflows, os agentes de desenvolvimento/produto e a base do
-installer do Wizz Method derivam do [BMAD Method](https://github.com/bmad-code-org/BMAD-METHOD),
-de BMad Code, LLC, licenciado sob MIT. Caminhos derivados: `src/bmm-skills/`,
-`src/core-skills/`, `src/modules/` (exceto `src/modules/wizz/`), `src/scripts/`
-e `tools/installer/`. Para esses caminhos, os termos MIT abaixo prevalecem
-sobre a LICENSE proprietária. O texto original da licença é reproduzido
-abaixo, como exigido pelos termos MIT.
+Wizz Method's workflow engine, development/product agents, and installer
+foundation derive from the [BMAD Method](https://github.com/bmad-code-org/BMAD-METHOD),
+by BMad Code, LLC, licensed under MIT. Derived paths: `src/bmm-skills/`,
+`src/core-skills/`, `src/modules/` (except `src/modules/wizz/`), `src/scripts/`,
+and `tools/installer/`. For those paths, the MIT terms below take
+precedence over the proprietary LICENSE. The original license text is
+reproduced below, as required by the MIT terms.
 
 ```
 MIT License
@@ -40,37 +40,37 @@ OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
 SOFTWARE.
 ```
 
-**Nota de marca:** BMad™, BMad Method™ e BMad Core™ são marcas de BMad Code,
-LLC, cobrindo todas as grafias e variações (incluindo BMAD, bmad, BMadMethod,
-BMAD-METHOD etc.). Essas marcas não são licenciadas junto com o código acima:
-a licença MIT cobre apenas o software. O Wizz Method não é endossado por
-BMad Code, LLC.
+**Trademark notice:** BMad™, BMad Method™, and BMad Core™ are trademarks of
+BMad Code, LLC, covering all spellings and variations (including BMAD, bmad,
+BMadMethod, BMAD-METHOD, etc.). These trademarks are not licensed along with
+the code above: the MIT license covers only the software. Wizz Method is not
+endorsed by BMad Code, LLC.
 
 ## wizz-ai-product-design (AI Product Design)
 
-- Caminho: `src/skills-lib/wizz-ai-product-design/references/`
-- Origem: [Owl-Listener/ai-design-skills](https://github.com/Owl-Listener/ai-design-skills)
-- Licença: MIT (texto completo em `src/skills-lib/wizz-ai-product-design/references/LICENSE`, copyright (c) 2026 MC Dean)
+- Path: `src/skills-lib/wizz-ai-product-design/references/`
+- Origin: [Owl-Listener/ai-design-skills](https://github.com/Owl-Listener/ai-design-skills)
+- License: MIT (full text at `src/skills-lib/wizz-ai-product-design/references/LICENSE`, copyright (c) 2026 MC Dean)
 
 ## wizz-algorithmic-art
 
-- Caminho: `src/skills-lib/wizz-algorithmic-art/`
-- Origem: skill de exemplo publicada pela Anthropic
-- Licença: Apache License 2.0 (texto completo em `src/skills-lib/wizz-algorithmic-art/LICENSE.txt`, copyright 2026 Anthropic, PBC)
+- Path: `src/skills-lib/wizz-algorithmic-art/`
+- Origin: example skill published by Anthropic
+- License: Apache License 2.0 (full text at `src/skills-lib/wizz-algorithmic-art/LICENSE.txt`, copyright 2026 Anthropic, PBC)
 
 ## wizz-canvas-design
 
-- Caminho: `src/skills-lib/wizz-canvas-design/`
-- Origem: skill de exemplo publicada pela Anthropic
-- Licença: Apache License 2.0 (texto completo em `src/skills-lib/wizz-canvas-design/LICENSE.txt`, copyright 2026 Anthropic, PBC)
+- Path: `src/skills-lib/wizz-canvas-design/`
+- Origin: example skill published by Anthropic
+- License: Apache License 2.0 (full text at `src/skills-lib/wizz-canvas-design/LICENSE.txt`, copyright 2026 Anthropic, PBC)
 
-### Fontes de `wizz-canvas-design/canvas-fonts/`
+### Fonts in `wizz-canvas-design/canvas-fonts/`
 
-As fontes abaixo são distribuídas sob a SIL Open Font License 1.1. O texto
-integral da OFL e o aviso de copyright de cada uma acompanham o próprio
-arquivo `<Fonte>-OFL.txt` na mesma pasta.
+The fonts below are distributed under the SIL Open Font License 1.1. The full
+OFL text and each font's copyright notice ship alongside the font itself, in
+the `<Font>-OFL.txt` file in the same folder.
 
-| Fonte | Autores / copyright |
+| Font | Authors / copyright |
 |---|---|
 | Arsenal SC | The Arsenal Project Authors |
 | Big Shoulders | The Big Shoulders Project Authors |
@@ -98,39 +98,39 @@ arquivo `<Fonte>-OFL.txt` na mesma pasta.
 | Silkscreen | The Silkscreen Project Authors |
 | Smooch Sans | The Smooch Sans Project Authors |
 | Tektur | The Tektur Project Authors |
-| Work Sans (inclui `WorkSans-BoldItalic.ttf`) | The Work Sans Project Authors |
+| Work Sans (includes `WorkSans-BoldItalic.ttf`) | The Work Sans Project Authors |
 | Young Serif | The Young Serif Project Authors |
 
 ## wizz-humanizer
 
-- Caminho: `src/skills-lib/wizz-humanizer/`
-- Origem: [blader/humanizer](https://github.com/blader/humanizer)
-- Licença: MIT (texto completo em `src/skills-lib/wizz-humanizer/LICENSE`, copyright (c) 2025 Siqi Chen)
+- Path: `src/skills-lib/wizz-humanizer/`
+- Origin: [blader/humanizer](https://github.com/blader/humanizer)
+- License: MIT (full text at `src/skills-lib/wizz-humanizer/LICENSE`, copyright (c) 2025 Siqi Chen)
 
 ## wizz-huashu-design
 
-- Caminho: `src/skills-lib/wizz-huashu-design/`
-- Origem: [alchaincyf/huashu-design](https://github.com/alchaincyf/huashu-design)
-- Licença: MIT (texto completo em `src/skills-lib/wizz-huashu-design/LICENSE`, copyright (c) 2026 alchaincyf (花叔))
+- Path: `src/skills-lib/wizz-huashu-design/`
+- Origin: [alchaincyf/huashu-design](https://github.com/alchaincyf/huashu-design)
+- License: MIT (full text at `src/skills-lib/wizz-huashu-design/LICENSE`, copyright (c) 2026 alchaincyf (花叔))
 
-## wizz-react-components (template Stitch)
+## wizz-react-components (Stitch template)
 
-- Caminho: `src/skills-lib/wizz-react-components/resources/component-template.tsx`,
+- Path: `src/skills-lib/wizz-react-components/resources/component-template.tsx`,
   `src/skills-lib/wizz-react-components/examples/gold-standard-card.tsx`,
   `src/skills-lib/wizz-react-components/scripts/fetch-stitch.sh`,
   `src/skills-lib/wizz-react-components/scripts/validate.js`
-- Origem: Google LLC (aviso de copyright no cabeçalho de cada arquivo)
-- Licença: Apache License 2.0 (texto completo em
+- Origin: Google LLC (copyright notice in each file's header)
+- License: Apache License 2.0 (full text at
   `src/skills-lib/wizz-react-components/LICENSE.txt`, copyright 2026 Google LLC)
 
 ## wizz-impeccable
 
-- Caminho: `src/skills-lib/wizz-impeccable/`
-- Origem: [pbakaus/impeccable](https://github.com/pbakaus/impeccable)
-- Licença: Apache License 2.0 (texto completo em `src/skills-lib/wizz-impeccable/LICENSE.txt`)
+- Path: `src/skills-lib/wizz-impeccable/`
+- Origin: [pbakaus/impeccable](https://github.com/pbakaus/impeccable)
+- License: Apache License 2.0 (full text at `src/skills-lib/wizz-impeccable/LICENSE.txt`)
 
 ## wizz-supabase-postgres-best-practices
 
-- Caminho: `src/skills-lib/wizz-supabase-postgres-best-practices/`
-- Origem: [supabase/agent-skills](https://github.com/supabase/agent-skills)
-- Licença: MIT (texto completo em `src/skills-lib/wizz-supabase-postgres-best-practices/LICENSE`, copyright (c) 2026 Supabase)
+- Path: `src/skills-lib/wizz-supabase-postgres-best-practices/`
+- Origin: [supabase/agent-skills](https://github.com/supabase/agent-skills)
+- License: MIT (full text at `src/skills-lib/wizz-supabase-postgres-best-practices/LICENSE`, copyright (c) 2026 Supabase)

package/package.json

@@ -1,7 +1,7 @@
 {
   "$schema": "https://json.schemastore.org/package.json",
   "name": "wizz-method",
-  "version": "2.1.1",
+  "version": "2.3.0",
   "description": "Wizz Method: método de agência orientado por IA em PT-BR",
   "keywords": [
     "agile",

package/src/bmm-skills/1-analysis/wizz-agent-analyst/customize.toml

@@ -37,7 +37,7 @@ activation_steps_append = []
 
 persistent_facts = [
   "file:{project-root}/**/project-context.md",
-  "FEATURE: o hook SessionStart já resolve a feature ativa a partir do branch git e injeta 'FEATURE ATIVA: <slug>'. Se ela existir, salve TODO artefato de planejamento e implementação na subpasta com esse nome ({planning_artifacts}/{active_feature}/, {implementation_artifacts}/{active_feature}/) e grave o slug em active_feature no {project-root}/_wizz/bmm/config.yaml na primeira escrita, avisando em UMA linha. Memória gravada nesta sessão leva o mesmo slug como tema. Sem feature resolvida, escreva nas raízes. NUNCA pergunte a feature ao usuário: invoque wizz-set-feature só se ele pedir para trocar, ou se houver duas features abertas e o pedido não decidir entre elas.",
+  "FEATURE: the SessionStart hook already resolves the active feature from the git branch and injects 'FEATURE ATIVA: <slug>'. If it exists, save EVERY planning and implementation artifact in the subfolder with that name ({planning_artifacts}/{active_feature}/, {implementation_artifacts}/{active_feature}/) and write the slug to active_feature in {project-root}/_wizz/bmm/config.yaml on first write, announcing it in ONE line. Memory saved in this session carries the same slug as its theme. With no feature resolved, write to the root folders. NEVER ask the user for the feature: invoke wizz-set-feature only if they ask to switch, or if there are two open features and the request does not decide between them.",
 ]
 
 role = "Help the user ideate research and analyze before committing to a project in the Wizz Method analysis phase."

package/src/bmm-skills/2-plan-workflows/wizz-agent-pm/customize.toml

@@ -37,7 +37,7 @@ activation_steps_append = []
 
 persistent_facts = [
   "file:{project-root}/**/project-context.md",
-  "FEATURE: o hook SessionStart já resolve a feature ativa a partir do branch git e injeta 'FEATURE ATIVA: <slug>'. Se ela existir, salve TODO artefato de planejamento e implementação na subpasta com esse nome ({planning_artifacts}/{active_feature}/, {implementation_artifacts}/{active_feature}/) e grave o slug em active_feature no {project-root}/_wizz/bmm/config.yaml na primeira escrita, avisando em UMA linha. Memória gravada nesta sessão leva o mesmo slug como tema. Sem feature resolvida, escreva nas raízes. NUNCA pergunte a feature ao usuário: invoque wizz-set-feature só se ele pedir para trocar, ou se houver duas features abertas e o pedido não decidir entre elas.",
+  "FEATURE: the SessionStart hook already resolves the active feature from the git branch and injects 'FEATURE ATIVA: <slug>'. If it exists, save EVERY planning and implementation artifact in the subfolder with that name ({planning_artifacts}/{active_feature}/, {implementation_artifacts}/{active_feature}/) and write the slug to active_feature in {project-root}/_wizz/bmm/config.yaml on first write, announcing it in ONE line. Memory saved in this session carries the same slug as its theme. With no feature resolved, write to the root folders. NEVER ask the user for the feature: invoke wizz-set-feature only if they ask to switch, or if there are two open features and the request does not decide between them.",
 ]
 
 role = "Translate product vision into a validated PRD, epics, and stories that development can execute during the Wizz Method planning phase."

package/src/bmm-skills/2-plan-workflows/wizz-agent-ux-designer/customize.toml

@@ -37,7 +37,7 @@ activation_steps_append = []
 
 persistent_facts = [
   "file:{project-root}/**/project-context.md",
-  "FEATURE: o hook SessionStart já resolve a feature ativa a partir do branch git e injeta 'FEATURE ATIVA: <slug>'. Se ela existir, salve TODO artefato de planejamento e implementação na subpasta com esse nome ({planning_artifacts}/{active_feature}/, {implementation_artifacts}/{active_feature}/) e grave o slug em active_feature no {project-root}/_wizz/bmm/config.yaml na primeira escrita, avisando em UMA linha. Memória gravada nesta sessão leva o mesmo slug como tema. Sem feature resolvida, escreva nas raízes. NUNCA pergunte a feature ao usuário: invoque wizz-set-feature só se ele pedir para trocar, ou se houver duas features abertas e o pedido não decidir entre elas.",
+  "FEATURE: the SessionStart hook already resolves the active feature from the git branch and injects 'FEATURE ATIVA: <slug>'. If it exists, save EVERY planning and implementation artifact in the subfolder with that name ({planning_artifacts}/{active_feature}/, {implementation_artifacts}/{active_feature}/) and write the slug to active_feature in {project-root}/_wizz/bmm/config.yaml on first write, announcing it in ONE line. Memory saved in this session carries the same slug as its theme. With no feature resolved, write to the root folders. NEVER ask the user for the feature: invoke wizz-set-feature only if they ask to switch, or if there are two open features and the request does not decide between them.",
 ]
 
 role = "Turn user needs and the PRD into UX design specifications that inform architecture and implementation during the Wizz Method planning phase."

package/src/bmm-skills/3-solutioning/wizz-agent-architect/customize.toml

@@ -37,7 +37,7 @@ activation_steps_append = []
 
 persistent_facts = [
   "file:{project-root}/**/project-context.md",
-  "FEATURE: o hook SessionStart já resolve a feature ativa a partir do branch git e injeta 'FEATURE ATIVA: <slug>'. Se ela existir, salve TODO artefato de planejamento e implementação na subpasta com esse nome ({planning_artifacts}/{active_feature}/, {implementation_artifacts}/{active_feature}/) e grave o slug em active_feature no {project-root}/_wizz/bmm/config.yaml na primeira escrita, avisando em UMA linha. Memória gravada nesta sessão leva o mesmo slug como tema. Sem feature resolvida, escreva nas raízes. NUNCA pergunte a feature ao usuário: invoque wizz-set-feature só se ele pedir para trocar, ou se houver duas features abertas e o pedido não decidir entre elas.",
+  "FEATURE: the SessionStart hook already resolves the active feature from the git branch and injects 'FEATURE ATIVA: <slug>'. If it exists, save EVERY planning and implementation artifact in the subfolder with that name ({planning_artifacts}/{active_feature}/, {implementation_artifacts}/{active_feature}/) and write the slug to active_feature in {project-root}/_wizz/bmm/config.yaml on first write, announcing it in ONE line. Memory saved in this session carries the same slug as its theme. With no feature resolved, write to the root folders. NEVER ask the user for the feature: invoke wizz-set-feature only if they ask to switch, or if there are two open features and the request does not decide between them.",
 ]
 
 role = "Convert the PRD and UX into technical architecture decisions that keep implementation on track during the Wizz Method solutioning phase."

package/src/bmm-skills/4-implementation/wizz-agent-dev/customize.toml

@@ -37,7 +37,7 @@ activation_steps_append = []
 
 persistent_facts = [
   "file:{project-root}/**/project-context.md",
-  "FEATURE: o hook SessionStart já resolve a feature ativa a partir do branch git e injeta 'FEATURE ATIVA: <slug>'. Se ela existir, salve TODO artefato de planejamento e implementação na subpasta com esse nome ({planning_artifacts}/{active_feature}/, {implementation_artifacts}/{active_feature}/) e grave o slug em active_feature no {project-root}/_wizz/bmm/config.yaml na primeira escrita, avisando em UMA linha. Memória gravada nesta sessão leva o mesmo slug como tema. Sem feature resolvida, escreva nas raízes. NUNCA pergunte a feature ao usuário: invoque wizz-set-feature só se ele pedir para trocar, ou se houver duas features abertas e o pedido não decidir entre elas.",
+  "FEATURE: the SessionStart hook already resolves the active feature from the git branch and injects 'FEATURE ATIVA: <slug>'. If it exists, save EVERY planning and implementation artifact in the subfolder with that name ({planning_artifacts}/{active_feature}/, {implementation_artifacts}/{active_feature}/) and write the slug to active_feature in {project-root}/_wizz/bmm/config.yaml on first write, announcing it in ONE line. Memory saved in this session carries the same slug as its theme. With no feature resolved, write to the root folders. NEVER ask the user for the feature: invoke wizz-set-feature only if they ask to switch, or if there are two open features and the request does not decide between them.",
   "When implementing against a library or framework, consult Context7 for current docs, APIs, and version-specific syntax before relying on training data or web search: call resolve-library-id with the library name, then query-docs with the returned /org/project id and a specific question. If Context7 is unavailable, tell the user once that they can enable it with `npx ctx7 setup`, then fall back to web research.",
 ]
 

package/src/bmm-skills/4-implementation/wizz-code-review/SKILL.md

@@ -1,11 +1,11 @@
 ---
 name: wizz-code-review
-description: 'Review code changes adversarially using parallel review layers (Blind Hunter, Edge Case Hunter, Verification Gap Reviewer, Acceptance Auditor) with structured triage into actionable categories. Use when the user says "run code review" or "review this code"'
+description: 'Adversarial code review in Forte-tier layers that scale with diff size, with triage into actionable categories. Use when the user says "run code review" or "review this code"'
 ---
 
 # Code Review Workflow
 
-> **Cross-reference:** this is the orchestrator. It gathers context, runs the review layers declared in `customize.toml` (Blind Hunter, Edge Case Hunter, Verification Gap Reviewer, Acceptance Auditor) as parallel context-free subagents driven by the prompt files under `review-prompts/`, then verifies, triages and presents findings by severity. Use it for a real diff/PR review. For a quick single-pass adversarial read with no workflow overhead, use `wizz-adversarial-reviewer` directly instead.
+> **Cross-reference:** this is the orchestrator. It gathers context, deterministically cleans the diff and picks a tier by its size, runs the review layers declared in `customize.toml` (Blind Hunter, Edge Case Hunter, Verification Gap Reviewer, Acceptance Auditor) as parallel context-free subagents — 1, 2 or up to 4 Forte-tier subagents depending on the tier — driven by the prompt files under `review-prompts/`, then verifies, triages on the Forte tier and presents findings by severity. Use it for a real diff/PR review. For a quick single-pass adversarial read with no workflow overhead, use `wizz-adversarial-reviewer` directly instead.
 
 **Goal:** Review code changes adversarially. No noise, no filler.