flyee 0.1.0

package/core/workflows/copy-collect.md

@@ -0,0 +1,306 @@
+---
+description:  Collect, validate, and normalize **textual content (copy)** required for pages, sections, components, or features **without mixing copy decisions with UI composition or design system logic**.
+---
+
+This workflow exists to:
+
+* Prevent copy decisions from being inferred implicitly
+* Separate **content authority** from **UI authority**
+* Ensure pages are never blocked by missing text
+* Maintain product and brand governance
+
+---
+
+## AUTHORITY & POSITION IN THE SYSTEM
+
+* `/copy-collect` is the **ONLY allowed workflow** to define or approve final copy
+* `/page-build` and `/ds-components` **MUST NOT invent final copy**
+* Copy is considered **product-level input**, not a UI decision
+
+---
+
+## HARD PRECONDITIONS
+
+Before execution, ALL conditions must be true:
+
+* `.agent/state/project.json` exists
+* `project.stage === "development"`
+* A target exists:
+
+  * page
+  * section
+  * component
+  * feature
+
+❌ If no target is defined → BLOCK execution
+
+---
+
+## INPUT
+
+### Required
+
+```text
+/copy-collect [target]
+```
+
+Examples:
+
+```
+/copy-collect onboarding
+/copy-collect dashboard
+/copy-collect page:settings
+/copy-collect component:pricing-card
+```
+
+---
+
+### Optional Inputs
+
+```text
+tone: formal | neutral | friendly | technical | marketing
+language: en | pt-BR | es | fr | custom
+brand_constraints: provided | unknown
+```
+
+---
+
+## COPY MODES (EXPLICIT)
+
+The workflow MUST determine which mode applies.
+
+### 1️⃣ Provided Copy (Preferred)
+
+User provides copy directly.
+
+```text
+Title:
+Subtitle:
+CTA:
+Descriptions:
+Empty states:
+Errors:
+```
+
+Rules:
+
+* Copy is stored verbatim
+* No rewriting unless explicitly requested
+* Considered FINAL unless reopened
+
+---
+
+### 2️⃣ Assisted Draft Copy (Controlled)
+
+If copy is not provided:
+
+* Generate **semantic placeholder copy**
+* Avoid brand tone assumptions
+* Avoid marketing language
+* Avoid claims or promises
+
+Allowed examples:
+
+```text
+"Primary action"
+"Section title"
+"Short descriptive text"
+```
+
+❌ Forbidden:
+
+* Value propositions
+* Emotional claims
+* Sales language
+* Feature naming
+
+All generated copy MUST be marked as:
+
+```text
+status: draft
+requires_approval: true
+```
+
+---
+
+### 3️⃣ Copy Approval Mode
+
+If draft copy already exists:
+
+* Present copy for approval
+* Allow:
+
+  * approve
+  * edit
+  * reject
+* No UI or page build may proceed with unapproved strategic copy
+
+---
+
+## WORKFLOW STEPS
+
+---
+
+### 1️⃣ Load Project Context
+
+Read:
+
+* `.agent/state/project.json`
+* Existing copy artifacts (if any)
+
+---
+
+### 2️⃣ Identify Copy Scope
+
+Determine required copy types:
+
+* Page titles
+* Section headers
+* Body text
+* CTAs
+* Labels
+* Empty states
+* Errors
+* Helper text
+* Accessibility labels (aria-labels)
+
+---
+
+### 3️⃣ Detect Copy Gaps
+
+For each required item:
+
+* exists → OK
+* missing → mark as REQUIRED
+
+---
+
+### 4️⃣ Resolve Copy Gaps
+
+Depending on mode:
+
+* Use provided copy
+* Generate placeholder copy
+* Request explicit user input
+
+---
+
+### 5️⃣ Normalize & Structure Copy
+
+Store copy in a structured format:
+
+```json
+{
+  "target": "dashboard",
+  "language": "en",
+  "tone": "neutral",
+  "status": "approved",
+  "copy": {
+    "title": "...",
+    "sections": [
+      {
+        "id": "overview",
+        "title": "...",
+        "description": "...",
+        "cta": "..."
+      }
+    ],
+    "empty_states": {},
+    "errors": {}
+  }
+}
+```
+
+---
+
+### 6️⃣ Persist Copy Artifact
+
+Write to:
+
+```
+src/content/
+└── copy/
+    └── [target].json
+```
+
+And register in:
+
+```json
+artifacts.documents[]
+```
+
+---
+
+### 7️⃣ Project State Update
+
+Update `.agent/state/project.json`:
+
+```json
+"content": {
+  "copy": {
+    "[target]": {
+      "status": "approved | draft",
+      "language": "en",
+      "last_updated": "ISO_TIMESTAMP"
+    }
+  }
+}
+```
+
+---
+
+## OUTPUT (MANDATORY)
+
+```text
+Copy Collection Status: COMPLETED
+Target: [page / component / feature]
+Copy Status: approved | draft
+Blocking Issues: none | approval required
+Next Allowed Action:
+- /page-build (if approved)
+- await copy approval
+```
+
+---
+
+## GOVERNANCE RULES (HARD LAW)
+
+* UI workflows MUST NOT invent final copy
+* Draft copy MUST be visibly marked
+* Strategic copy requires explicit approval
+* Copy changes DO NOT invalidate Design System validation
+* Copy does NOT unlock UI by itself
+
+---
+
+## /next-step INTEGRATION (MANDATORY)
+
+`/next-step` MUST enforce:
+
+```text
+If page-build is requested AND copy.status !== "approved":
+→ Recommend /copy-collect
+```
+
+---
+
+## RELATIONSHIP TO OTHER WORKFLOWS
+
+| Workflow         | Relationship                  |
+| ---------------- | ----------------------------- |
+| `/page-build`    | Consumes approved copy        |
+| `/ds-components` | May use labels only           |
+| `/ds-validate`   | Ignores copy (structure only) |
+| `/next-step`     | Orchestrates order            |
+| `/copy-collect`  | Content authority             |
+
+---
+
+## GOLDEN RULE
+
+> **Design defines structure.
+> Copy defines meaning.
+> Data defines truth.
+> Governance defines order.**
+
+---

package/core/workflows/create-agent.md

@@ -0,0 +1,33 @@
+---
+description: Create new specialist agent
+---
+
+# /create-agent
+
+> Interactively create a new specialist agent. Can invoke /create-skill if needed.
+
+## 1. Identity
+1.  **Identity**: "What is the agent's name?" (e.g., `aws-specialist`, `tanavitrine-expert`)
+2.  **Trigger**: "What keywords should trigger this agent?"
+
+> **Note:** All agents are created in `.agent/agents/` (global scope). The `.agent/` directory is immutable per project.
+
+## 2. Skills Analysis
+1.  **Existing**: "Which existing skills should this agent possess?" (List common ones like `clean-code`)
+2.  **New**: "Do you need a new specific skill?"
+    - If YES: Invoke `/create-skill` first.
+
+## 3. Behavior Definition
+1.  **Philosophy**: "What is this agent's core philosophy?"
+2.  **Do/Don'ts**: "What are the critical Do's and Don'ts?"
+3.  **Questions**: "What must the agent ask before acting?"
+
+## 4. Execution
+1.  **Target Directory**: `.agent/agents/`
+2.  **Read Template**: `.agent/templates/agent-template.md`
+3.  **Generate File**: Write `{name}.md` in the target directory.
+
+## Example
+```bash
+/create-agent --local
+```

package/core/workflows/create-skill.md

@@ -0,0 +1,39 @@
+---
+description: Create new skill with guided dialogue
+---
+
+# /create-skill
+
+> Interactively create a new specialized skill in global or local scope.
+
+## 1. Domain (Socratic)
+First, determine the skill's purpose:
+1.  **Domain**: "What is the specific domain or technology?" (e.g., "GraphQL Patterns", "Vitrine Rules")
+2.  **Name**: "What should be the folder name?" (kebab-case, e.g., `graphql-patterns`)
+
+> **Note:** All skills are created in `.agent/skills/` (global scope). The `.agent/` directory is immutable per project.
+
+## 2. Structure
+Ask about the skill's components:
+- "Does this skill need auxiliary scripts?"
+- "Does it need templates or examples?"
+
+## 3. Content Generation
+Ask for core content to populate the template:
+- "What are the top 3 core principles?"
+- "What are the critical anti-patterns?"
+- "Are there specific decision frameworks?"
+
+## 4. Execution
+1.  **Target Directory**: `.agent/skills/{name}`
+2.  **Read Template**: `.agent/templates/skill-template.md`
+3.  **Generate File**: Write `SKILL.md` in the target directory with populated content.
+4.  **Optional**: 
+    - If scripts requested: Create `scripts/` folder and copy `.agent/templates/script-template.py` to `scripts/verify.py`.
+    - If examples requested: Create `examples/` folder.
+
+## Example
+```bash
+/create-skill
+# Follow the dialogue
+```

package/core/workflows/create-workflow.md

@@ -0,0 +1,33 @@
+---
+description: Create new workflow (can invoke create-agent)
+---
+
+# /create-workflow
+
+> Interactively create a new workflow. Can invoke /create-agent if a new specialist is needed.
+
+## 1. Purpose
+1.  **Trigger**: "What is the slash command?" (e.g., `/deploy-aws`)
+2.  **Goal**: "What problem does this workflow solve?"
+
+> **Note:** All workflows are created in `.agent/workflows/` (global scope). The `.agent/` directory is immutable per project.
+
+## 2. Agent Check
+1.  **Required expertise**: "Which agents are needed to execute this?"
+2.  **Gap Analysis**: "Do these agents exist?"
+    - If NO: Invoke `/create-agent` to create the missing specialist.
+
+## 3. Flow Design
+1.  **Phases**: "What are the main phases?"
+2.  **Steps**: "List the key steps for each phase."
+3.  **Inputs/Outputs**: "What inputs are needed? What artifacts are produced?"
+
+## 4. Execution
+1.  **Target Directory**: `.agent/workflows/`
+2.  **Read Template**: `.agent/templates/workflow-template.md`
+3.  **Generate File**: Write `{command}.md` in the target directory.
+
+## Example
+```bash
+/create-workflow
+```

package/core/workflows/create.md

@@ -0,0 +1,92 @@
+---
+description: Create new application command. Triggers App Builder skill and starts interactive dialogue with user.
+---
+
+# /create - Create Application
+
+$ARGUMENTS
+
+---
+
+## Task
+
+This command starts a new application creation process.
+
+### Steps:
+
+1. **Request Analysis**
+   - Understand what the user wants
+   - If information is missing, use `conversation-manager` skill to ask
+
+2. **Project Planning**
+   - Use `project-planner` agent for task breakdown
+   - Determine tech stack
+   - Plan file structure
+   - Create plan file and proceed to building
+
+3. **TDD Metodologia (OBRIGATÓRIO)**
+   > [!IMPORTANT]
+   > **Antes de implementar, escrever testes.**
+   
+   - Use `tdd-workflow` skill para cada feature
+   - Ciclo: 🔴 RED → 🟢 GREEN → 🔵 REFACTOR
+   - Gerar testes com `/test [feature]`
+
+4. **Design System (OBRIGATÓRIO para UI)**
+   > [!IMPORTANT]
+   > **SKILL OBRIGATÓRIA:** Seguir `ui-ux-discovery` para perguntas granulares.
+   > **WORKFLOW:** Executar `/ui-ux-pro-max` para obter recomendações.
+   
+   **Processo:**
+   1. Executar script para obter recomendações
+   2. **Fazer perguntas granulares** (cores, tipografia, layout, efeitos, logo)
+   3. **AGUARDAR respostas** antes de finalizar
+   4. Consolidar decisões e persistir em `design-system/MASTER.md`
+   5. AGUARDAR aprovação antes de implementar
+
+5. **Application Building (After Tests + Design System + Approval)**
+   - Orchestrate with `app-builder` skill
+   - Coordinate expert agents:
+     - `database-architect` → Schema
+     - `backend-specialist` → API
+     - `frontend-specialist` → UI (seguindo Design System)
+
+6. **Verificação de Cobertura (GATE)**
+   > [!CAUTION]
+   > **Cobertura mínima: 80%** antes de preview.
+   
+   ```bash
+   /test coverage
+   ```
+   
+   Se cobertura < 80%:
+   - Identificar áreas não cobertas
+   - Adicionar testes faltantes
+   - Repetir verificação
+
+7. **Preview**
+   - Start with `auto_preview.py` when complete
+   - Present URL to user
+
+---
+
+## Usage Examples
+
+```
+/create blog site
+/create e-commerce app with product listing and cart
+/create todo app
+/create Instagram clone
+/create crm system with customer management
+```
+
+---
+
+## Before Starting
+
+If request is unclear, ask these questions:
+- What type of application?
+- What are the basic features?
+- Who will use it?
+
+Use defaults, add details later.

package/core/workflows/debug.md

@@ -0,0 +1,186 @@
+---
+description: Debugging command. Activates DEBUG mode for systematic problem investigation.
+---
+
+# /debug - Systematic Problem Investigation
+
+$ARGUMENTS
+
+---
+
+## Purpose
+
+This command activates DEBUG mode for systematic investigation of issues, errors, or unexpected behavior.
+
+---
+
+## Agentes Envolvidos
+
+- `debugger` - Investigação sistemática e análise de root cause
+- `backend-specialist` / `frontend-specialist` / `mobile-developer` - Conforme domínio do erro
+- `test-engineer` - Para criar testes de regressão após correção
+
+---
+
+## Etapa 0: Flyee Sync (OBRIGATÓRIO — ANTES de qualquer código)
+
+> [!CAUTION]
+> Toda investigação de bug DEVE ser registrada no Flyee antes de iniciar qualquer fix.
+> Se `flyee.json` não existir ou `enabled: false`, informar o usuário.
+
+```bash
+python3 .agent/flyee-bridge/bridge.py --create-task \
+  --name "Bug: <descrição curta do problema>" \
+  --type generic \
+  --description "<sintoma observado, arquivo afetado, passos para reproduzir>" \
+  --priority high
+```
+
+> Salvar o `task_id` retornado — será usado nas etapas seguintes.
+
+---
+
+## Etapa 1: Gather Information
+
+- Error message / sintoma visível
+- Reproduction steps
+- Expected vs actual behavior
+- Recent changes that may have caused it
+
+---
+
+## Etapa 2: Form Hypotheses
+
+- List possible causes
+- Order by likelihood
+
+---
+
+## Etapa 3: Investigate Systematically
+
+- Test each hypothesis
+- Check logs, data flow
+- Use elimination method
+
+---
+
+## Etapa 4: Fix and Prevent
+
+- Apply fix
+- Explain root cause
+- Add prevention measures
+
+---
+
+## Etapa 5: QA Test Gate (BLOQUEANTE — ANTES de marcar concluído)
+
+> [!CAUTION]
+> O bug NÃO pode ser marcado como concluído sem gerar e executar testes.
+> Seguir heurísticas de `@[skills/qa-test-generation]`.
+
+1. **Transicionar status para `testing`:**
+```bash
+python3 .agent/flyee-bridge/bridge.py --update-task <task_id> --status testing
+```
+
+2. **Gerar checklist de testes:**
+```bash
+python3 .agent/flyee-bridge/bridge.py --generate-tests <task_id>
+```
+
+3. **Rodar testes automáticos** (TypeScript, Vitest, Playwright — conforme stack):
+```bash
+python3 .agent/flyee-bridge/bridge.py --report-test <task_id> <step_id> passed|failed ["comment"]
+```
+
+4. **Verificar gate:**
+```bash
+python3 .agent/flyee-bridge/bridge.py --test-summary <task_id>
+```
+
+- `all_passed == true` → ✅ prosseguir para Etapa 6
+- `all_passed == false` → ❌ BLOQUEAR — corrigir testes antes de concluir
+
+> Se o usuário solicitar skip (ex: hotfix urgente), registrar como `skipped` com aviso de quality debt.
+
+> [!IMPORTANT]
+> **Para bugs de UI:** orientar o usuário a reproduzir o bug pelo caminho real da aplicação.
+> Exemplo: abrir a tela de tasks de um projeto → clicar nos 3 pontos de uma task → selecionar uma opção que abre modal → fechar o modal → verificar se a tela continua responsiva.
+
+---
+
+## Etapa 6: Conclusão — /task-complete
+
+> Após testes aprovados, executar obrigatoriamente:
+
+```
+/task-complete <task_id> "<tempo_gasto>"
+```
+
+Isso sincroniza o status, adiciona nota de conclusão e comentário rico no Flyee.
+
+---
+
+## Output Format
+
+```markdown
+## 🔍 Debug: [Issue]
+
+### 1. Symptom
+[What's happening]
+
+### 2. Information Gathered
+- Error: `[error message]`
+- File: `[filepath]`
+- Line: [line number]
+
+### 3. Hypotheses
+1. ❓ [Most likely cause]
+2. ❓ [Second possibility]
+3. ❓ [Less likely cause]
+
+### 4. Investigation
+
+**Testing hypothesis 1:**
+[What I checked] → [Result]
+
+**Testing hypothesis 2:**
+[What I checked] → [Result]
+
+### 5. Root Cause
+🎯 **[Explanation of why this happened]**
+
+### 6. Fix
+```[language]
+// Before
+[broken code]
+
+// After
+[fixed code]
+```
+
+### 7. Prevention
+🛡️ [How to prevent this in the future]
+```
+
+---
+
+## Examples
+
+```
+/debug login not working
+/debug API returns 500
+/debug form doesn't submit
+/debug modal freezes the screen
+```
+
+---
+
+## Key Principles
+
+- **Register first** - create Flyee task before touching code
+- **Ask before assuming** - get full error context
+- **Test hypotheses** - don't guess randomly
+- **Explain why** - not just what to fix
+- **Prevent recurrence** - add tests, validation
+- **No completion without tests** - QA gate is mandatory