mega-brain-ai 1.1.0

package/.claude/skills/plugin-dev/SKILL.md

@@ -0,0 +1,176 @@
+# Plugin Development Skill
+
+> **Auto-Trigger:** Quando usuário quiser criar plugin, desenvolver extensão, criar skill customizada
+> **Keywords:** "plugin", "desenvolver plugin", "criar plugin", "extensão", "criar skill", "plugin-dev", "create-plugin"
+> **Prioridade:** ALTA
+> **Namespace:** [OFFICIAL]
+> **Tools:** Read, Write, Edit, Glob, Grep, Bash, Task
+
+## Quando NÃO Ativar
+
+- Uso de plugins existentes sem modificação
+- Debugging de plugins já implementados
+- Perguntas gerais sobre plugins
+- Quando usuário explicitamente pedir skill diferente
+
+---
+
+## Core Purpose
+
+Comprehensive toolkit for developing Claude Code plugins. Provides structured 8-phase
+workflow from concept to deployment with 7 core skills for plugin creation.
+
+---
+
+## The 7 Core Skills
+
+| Skill | Purpose |
+|-------|---------|
+| **Architecture** | Design plugin structure and interfaces |
+| **Implementation** | Write plugin code following best practices |
+| **Testing** | Create comprehensive test suites |
+| **Documentation** | Write clear plugin documentation |
+| **Integration** | Connect with Claude Code systems |
+| **Optimization** | Improve performance and reliability |
+| **Deployment** | Package and distribute plugins |
+
+---
+
+## The 8 Phases of Plugin Creation
+
+### Phase 1: Concept
+
+- Define plugin purpose
+- Identify target users
+- List core features
+- Set success criteria
+
+### Phase 2: Research
+
+- Study similar plugins
+- Review Claude Code plugin API
+- Identify dependencies
+- Note limitations
+
+### Phase 3: Design
+
+- Create plugin architecture
+- Define file structure
+- Design interfaces
+- Plan extension points
+
+### Phase 4: Scaffold
+
+- Create directory structure
+- Initialize configuration files
+- Set up development environment
+- Create boilerplate code
+
+### Phase 5: Implementation
+
+- Write core functionality
+- Implement features incrementally
+- Handle errors gracefully
+- Follow coding standards
+
+### Phase 6: Testing
+
+- Write unit tests
+- Create integration tests
+- Test edge cases
+- Verify error handling
+
+### Phase 7: Documentation
+
+- Write README
+- Document API/interfaces
+- Create usage examples
+- Add troubleshooting guide
+
+### Phase 8: Deployment
+
+- Package plugin
+- Version appropriately
+- Publish/distribute
+- Monitor feedback
+
+---
+
+## Plugin Structure Template
+
+```
+plugin-name/
+├── SKILL.md           # Main skill definition
+├── README.md          # Documentation
+├── package.json       # If npm-based
+├── src/
+│   ├── index.ts       # Entry point
+│   ├── handlers/      # Event handlers
+│   └── utils/         # Helper functions
+├── tests/
+│   └── *.test.ts      # Test files
+└── examples/
+    └── *.md           # Usage examples
+```
+
+---
+
+## SKILL.md Template
+
+```markdown
+# [Plugin Name] Skill
+
+> **Auto-Trigger:** [When to activate]
+> **Keywords:** "keyword1", "keyword2", "keyword3"
+> **Prioridade:** [ALTA | MÉDIA | BAIXA]
+> **Tools:** [List of tools used]
+
+## Quando NÃO Ativar
+
+- [Situation 1]
+- [Situation 2]
+
+---
+
+## Core Purpose
+
+[Brief description of what the plugin does]
+
+---
+
+## [Main Content Sections]
+
+[Plugin-specific documentation]
+
+---
+
+## Example Prompts This Skill Handles
+
+- "[Example 1]"
+- "[Example 2]"
+```
+
+---
+
+## Command
+
+Use `/plugin-dev:create-plugin` to start the guided plugin creation workflow.
+
+---
+
+## Example Prompts This Skill Handles
+
+- "Crie um plugin para processar PDFs"
+- "Develop a plugin for database migrations"
+- "Quero criar uma skill para análise de logs"
+- "Build an extension for API testing"
+- "Desenvolva um plugin de formatação de código"
+
+---
+
+## Integration Notes
+
+This skill is part of the **Claude Code Official Skills** collection.
+Namespace: `[OFFICIAL]` - Distinguishes from Mega Brain custom skills.
+
+For custom Mega Brain skills, see: `/.claude/skills/`

package/.claude/skills/pr-review-toolkit/SKILL.md

@@ -0,0 +1,178 @@
+# PR Review Toolkit Skill
+
+> **Auto-Trigger:** Quando usuário pedir análise de PR, revisão de pull request, análise de comentários, hunting de falhas
+> **Keywords:** "PR", "pull request", "revisar PR", "análise de PR", "PR review", "comment analyzer", "silent failure"
+> **Prioridade:** ALTA
+> **Namespace:** [OFFICIAL]
+> **Tools:** Read, Glob, Grep, Bash, Task
+
+## Quando NÃO Ativar
+
+- Code review geral sem contexto de PR
+- Criação de PRs (usar git workflow)
+- Perguntas sobre PRs sem análise
+- Quando usuário explicitamente pedir skill diferente
+
+---
+
+## Core Purpose
+
+Bundle of 6 specialized agents for comprehensive PR analysis. Provides deep inspection
+of comments, tests, silent failures, type design, code quality, and simplification opportunities.
+
+---
+
+## The 6 Specialized Agents
+
+### 1. Comment Analyzer
+
+**Focus:** PR comment triage
+
+- Categorizes comments by severity
+- Identifies blocking vs non-blocking
+- Summarizes discussion threads
+- Highlights unresolved items
+- Tracks comment resolution status
+
+### 2. PR Test Analyzer
+
+**Focus:** Test coverage validation
+
+- Identifies untested code paths
+- Suggests missing test cases
+- Validates test quality
+- Checks edge case coverage
+- Reviews test naming/structure
+
+### 3. Silent Failure Hunter
+
+**Focus:** Hidden bug detection
+
+- Finds swallowed exceptions
+- Identifies missing error handling
+- Detects async/await issues
+- Spots race conditions
+- Finds resource leaks
+
+### 4. Type Design Analyzer
+
+**Focus:** Type system quality
+
+- Reviews type definitions
+- Identifies type safety issues
+- Suggests better type patterns
+- Checks generic usage
+- Validates interface design
+
+### 5. Code Reviewer
+
+**Focus:** General code quality
+
+- Style consistency
+- Best practices adherence
+- Pattern compliance
+- Documentation quality
+- Naming conventions
+
+### 6. Code Simplifier
+
+**Focus:** Complexity reduction
+
+- Identifies over-engineering
+- Suggests simplifications
+- Removes dead code
+- Consolidates duplicates
+- Improves readability
+
+---
+
+## Agent Activation
+
+Each agent can be invoked individually or as a full suite:
+
+| Command | Agents Activated |
+|---------|------------------|
+| `/pr-review:full` | All 6 agents |
+| `/pr-review:comments` | Comment Analyzer |
+| `/pr-review:tests` | PR Test Analyzer |
+| `/pr-review:failures` | Silent Failure Hunter |
+| `/pr-review:types` | Type Design Analyzer |
+| `/pr-review:quality` | Code Reviewer |
+| `/pr-review:simplify` | Code Simplifier |
+
+---
+
+## Output Format
+
+```
+╔══════════════════════════════════════════════════════════════════════════════╗
+║                         PR REVIEW TOOLKIT REPORT                             ║
+╠══════════════════════════════════════════════════════════════════════════════╣
+║  PR: #[number] - [title]                                                     ║
+║  Agents Run: [list]                                                          ║
+║  Files Analyzed: [N]                                                         ║
+╚══════════════════════════════════════════════════════════════════════════════╝
+
+┌─ COMMENT ANALYZER ──────────────────────────────────────────────────────────┐
+│  Blocking: [N]  |  Suggestions: [N]  |  Resolved: [N]                       │
+│                                                                             │
+│  BLOCKING ITEMS:                                                            │
+│  - [item description] @ file:line                                           │
+└─────────────────────────────────────────────────────────────────────────────┘
+
+┌─ SILENT FAILURE HUNTER ─────────────────────────────────────────────────────┐
+│  Potential Issues Found: [N]                                                │
+│                                                                             │
+│  ⚠️  [issue description]                                                    │
+│      Location: file:line                                                    │
+│      Risk: [HIGH | MEDIUM | LOW]                                            │
+└─────────────────────────────────────────────────────────────────────────────┘
+
+┌─ TEST ANALYZER ─────────────────────────────────────────────────────────────┐
+│  Coverage Delta: [+/-X%]                                                    │
+│  Missing Tests: [N]                                                         │
+│                                                                             │
+│  SUGGESTED TESTS:                                                           │
+│  - [test case description]                                                  │
+└─────────────────────────────────────────────────────────────────────────────┘
+
+┌─ TYPE DESIGN ANALYZER ──────────────────────────────────────────────────────┐
+│  Type Issues: [N]                                                           │
+│                                                                             │
+│  RECOMMENDATIONS:                                                           │
+│  - [recommendation]                                                         │
+└─────────────────────────────────────────────────────────────────────────────┘
+
+┌─ CODE SIMPLIFIER ───────────────────────────────────────────────────────────┐
+│  Simplification Opportunities: [N]                                          │
+│                                                                             │
+│  SUGGESTIONS:                                                               │
+│  - [simplification suggestion]                                              │
+└─────────────────────────────────────────────────────────────────────────────┘
+
+┌─ SUMMARY ───────────────────────────────────────────────────────────────────┐
+│  ✅ Ready to merge: [YES | NO | NEEDS WORK]                                 │
+│  📋 Action items: [N]                                                       │
+│  ⏱️  Estimated fix time: [estimate]                                         │
+└─────────────────────────────────────────────────────────────────────────────┘
+```
+
+---
+
+## Example Prompts This Skill Handles
+
+- "Analise o PR #123 completamente"
+- "Review the pull request for silent failures"
+- "Verifique os comentários não resolvidos no PR"
+- "Hunt for async issues in this PR"
+- "Simplifique o código deste pull request"
+- "Check test coverage for PR changes"
+
+---
+
+## Integration Notes
+
+This skill is part of the **Claude Code Official Skills** collection.
+Namespace: `[OFFICIAL]` - Distinguishes from Mega Brain custom skills.
+
+For custom Mega Brain skills, see: `/.claude/skills/`

package/.claude/skills/resume/SKILL.md

@@ -0,0 +1,61 @@
+# RESUME SESSION - Recuperação de Contexto
+
+## Trigger
+`/resume` ou ao iniciar nova conversa
+
+## Objetivo
+Recuperar o contexto COMPLETO da última sessão para continuar de onde parou.
+
+## Execução
+
+### 1. Localizar Última Sessão
+Ler `.claude/sessions/LATEST-SESSION.md` para encontrar a sessão mais recente.
+
+### 2. Carregar Contexto
+Ler o arquivo de sessão completo e extrair:
+- Estado da missão
+- Fase atual
+- Pendências
+- Próximos passos
+- Decisões tomadas
+- Notas importantes
+
+### 3. Apresentar Resumo de Retomada
+
+```
+┌──────────────────────────────────────────────────────────────────────────────┐
+│  🔄 RETOMANDO SESSÃO                                                         │
+│                                                                              │
+│  Última atividade: [DATA/HORA]                                              │
+│  Duração desde última sessão: [X horas/dias]                                │
+│                                                                              │
+│  📍 ONDE PARAMOS:                                                            │
+│  Missão: [NOME]                                                              │
+│  Fase: [N] de 5 - [NOME_FASE]                                               │
+│  Progresso: [X]%                                                             │
+│                                                                              │
+│  📋 PENDÊNCIAS HERDADAS:                                                     │
+│  - [Pendência 1]                                                             │
+│  - [Pendência 2]                                                             │
+│                                                                              │
+│  ➡️ PRÓXIMO PASSO PLANEJADO:                                                 │
+│  [Descrição do próximo passo]                                                │
+│                                                                              │
+│  💡 DECISÕES ANTERIORES:                                                     │
+│  - [Decisão relevante]                                                       │
+│                                                                              │
+└──────────────────────────────────────────────────────────────────────────────┘
+```
+
+### 4. Perguntar Confirmação
+"Quer continuar de onde paramos ou precisa de algo diferente?"
+
+### 5. Listar Sessões Antigas (Opcional)
+Se usuário pedir, listar todas as sessões disponíveis:
+```
+/resume list - mostra todas as sessões salvas
+/resume [SESSION-ID] - carrega sessão específica
+```
+
+## Output
+Contexto recuperado + resumo visual + confirmação do usuário

package/.claude/skills/save/SKILL.md

@@ -0,0 +1,87 @@
+# SAVE SESSION - Persistência de Contexto
+
+## Trigger
+`/save` ou automaticamente ao final de cada tarefa significativa
+
+## Objetivo
+Salvar o estado COMPLETO da sessão atual para recuperação posterior.
+
+## Execução
+
+Quando acionado:
+
+### 1. Coletar Estado Atual
+```
+- Timestamp da sessão
+- Missão ativa (se houver)
+- Fase atual
+- Últimas ações executadas
+- Arquivos modificados
+- Pendências identificadas
+- Contexto de conversa (resumo estruturado)
+```
+
+### 2. Gerar SESSION-LOG
+Criar arquivo em `.claude/sessions/` com formato:
+```
+SESSION-YYYY-MM-DD-HHmm.md
+```
+
+### 3. Template Obrigatório
+
+```markdown
+# SESSION LOG - [TIMESTAMP]
+
+## ESTADO DA MISSÃO
+- **Missão**: [NOME]
+- **Fase**: [N] de 5
+- **Progresso**: [X]%
+
+## CONTEXTO DA CONVERSA
+[Resumo detalhado do que foi discutido/decidido]
+
+## AÇÕES EXECUTADAS
+1. [Ação 1 com detalhes]
+2. [Ação 2 com detalhes]
+...
+
+## ARQUIVOS MODIFICADOS
+- [arquivo1.ext] - [o que mudou]
+- [arquivo2.ext] - [o que mudou]
+
+## PENDÊNCIAS
+- [ ] [Pendência 1]
+- [ ] [Pendência 2]
+
+## DECISÕES TOMADAS
+- [Decisão 1]: [Razão]
+- [Decisão 2]: [Razão]
+
+## PRÓXIMOS PASSOS PLANEJADOS
+1. [Próximo passo 1]
+2. [Próximo passo 2]
+
+## NOTAS IMPORTANTES
+[Qualquer coisa crítica para não esquecer]
+
+---
+Session ID: [UUID]
+Saved at: [ISO timestamp]
+```
+
+### 4. Atualizar LATEST-SESSION.md
+Sempre manter um link para a última sessão em:
+`.claude/sessions/LATEST-SESSION.md`
+
+### 5. Confirmar
+Responder: "Sessão salva: SESSION-YYYY-MM-DD-HHmm.md"
+
+## Auto-Save Triggers
+- Após completar qualquer batch
+- Antes de qualquer operação destrutiva
+- A cada 30 minutos de atividade
+- Quando detectar pausa prolongada
+- Quando usuário expressar que vai sair
+
+## Output
+Arquivo de sessão + confirmação visual

package/.claude/skills/skill-writer/SKILL.md

@@ -0,0 +1,153 @@
+---
+name: skill-writer
+description: Guide users through creating Agent Skills for Claude Code. Use when the user wants to create, write, author, or design a new Skill, or needs help with SKILL.md files, frontmatter, or skill structure.
+allowed-tools: Read, Grep, Glob, Write, Edit, Bash(mkdir:*), Bash(ls:*)
+---
+
+# Skill Writer
+
+Create well-structured Agent Skills for Claude Code following best practices.
+
+## Quick start
+
+```yaml
+---
+name: my-skill-name
+description: What this does. Use when user mentions X, Y, or Z.
+---
+
+# My Skill Name
+
+Instructions for Claude here...
+```
+
+## When to use this Skill
+
+- Creating a new Agent Skill
+- Writing or updating SKILL.md files
+- Designing skill structure and frontmatter
+- Converting existing prompts into Skills
+
+## Instructions
+
+### Step 1: Determine scope
+
+Ask clarifying questions:
+- What specific capability should this Skill provide?
+- When should Claude use this Skill?
+- What tools or resources does it need?
+
+**Rule**: One Skill = one capability
+
+### Step 2: Choose location
+
+| Location | Use Case |
+|----------|----------|
+| `~/.claude/skills/` | Personal, experimental |
+| `.claude/skills/` | Team/project, git-tracked |
+
+### Step 3: Create structure
+
+```bash
+mkdir -p .claude/skills/skill-name
+```
+
+Multi-file structure:
+```
+skill-name/
+├── SKILL.md (required)
+├── reference.md (optional)
+├── examples.md (optional)
+└── scripts/ (optional)
+```
+
+### Step 4: Write frontmatter
+
+```yaml
+---
+name: skill-name
+description: Brief description of what this does and when to use it
+---
+```
+
+| Field | Rules |
+|-------|-------|
+| `name` | Lowercase, hyphens, max 64 chars, **must match folder name** |
+| `description` | Max 1024 chars, include WHAT + WHEN to use |
+| `allowed-tools` | (Optional) Restrict tool access |
+
+### Step 5: Write effective descriptions
+
+**Formula**: `[What it does] + [When to use it] + [Key triggers]`
+
+```yaml
+# Good
+description: Extract text from PDFs, fill forms. Use when working with PDF files or document extraction.
+
+# Bad
+description: Helps with documents
+```
+
+Tips:
+- Include file extensions (.pdf, .xlsx)
+- Use trigger words ("analyze", "extract", "generate")
+- Add "Use when..." clause
+
+### Step 6: Structure content
+
+```markdown
+# Skill Name
+
+Brief overview.
+
+## Quick start
+
+Simple example.
+
+## Instructions
+
+Step-by-step guidance.
+
+## Requirements
+
+Dependencies if any.
+```
+
+### Step 7: Validate
+
+- [ ] SKILL.md exists in correct location
+- [ ] Directory name matches frontmatter `name`
+- [ ] YAML valid (no tabs, spaces only)
+- [ ] Description < 1024 chars, includes what + when
+- [ ] Instructions are actionable
+
+### Step 8: Test
+
+1. Restart Claude Code to load Skill
+2. Ask questions matching the description
+3. Verify activation and behavior
+
+## Additional resources
+
+- **Patterns and examples**: See [examples.md](examples.md)
+- **Debugging issues**: See [troubleshooting.md](troubleshooting.md)
+
+## Output format
+
+When creating a Skill, I will:
+
+1. Ask clarifying questions about scope
+2. Suggest name and location
+3. Create SKILL.md with proper frontmatter
+4. Include instructions and examples
+5. Add supporting files if needed
+6. Validate against requirements
+
+## Best practices summary
+
+1. **One Skill, one purpose**
+2. **Specific descriptions** with trigger words
+3. **Clear instructions** written for Claude
+4. **Concrete examples** with real code
+5. **Progressive disclosure** for complex Skills
+6. **Under 500 lines** in SKILL.md