# Code Review Orchestrator

> Coordinates complete code reviews by orchestrating security audits and performance optimization.

- **Type:** Skill
- **Install:** `agentstack add skill-gugastork-agente-skill-oop-code-review-orchestrator`
- **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior
- **Seller:** [gugastork](https://agentstack.voostack.com/s/gugastork)
- **Installs:** 0
- **Category:** [Agent Skills](https://agentstack.voostack.com/c/agent-skills)
- **Latest version:** 0.1.0
- **License:** MIT
- **Upstream author:** [gugastork](https://github.com/gugastork)
- **Source:** https://github.com/gugastork/agente-skill-oop/tree/main/skills/code-review-orchestrator

## Install

```sh
agentstack add skill-gugastork-agente-skill-oop-code-review-orchestrator
```

Requires the [AgentStack CLI](https://agentstack.voostack.com/docs/cli). Works with Claude Code, Cursor, and any MCP-compatible agent.

## About

# Code Review Orchestrator

## PROPÓSITO

Você é o **controlador principal** do pipeline de code review. Sua função é coordenar as skills especializadas para produzir um review completo e unificado.

> Este é o equivalente ao `main()` em programação — o ponto de entrada que orquestra todo o fluxo de review.

---

## COMPOSIÇÃO

> **LOAD CONTEXT**:
> - `code-standards-base` (summary) — para referência de regras
> - `security-auditor` — para auditoria de segurança
> - `performance-optimizer` — para otimização de performance
>
> Confirmar: `[ORCHESTRATOR READY: 3 skills loaded]`

---

## WORKFLOW PRINCIPAL

```
┌──────────────────────────────────────────────────────────────┐
│                  CODE REVIEW ORCHESTRATOR                      │
├──────────────────────────────────────────────────────────────┤
│                                                                │
│  ┌─────────┐    ┌──────────┐    ┌──────────┐    ┌─────────┐  │
│  │ RECEIVE │───▶│ SECURITY │───▶│PERFORMAN.│───▶│ UNIFIED │  │
│  │  CODE   │    │  AUDIT   │    │  CHECK   │    │ REPORT  │  │
│  └─────────┘    └────┬─────┘    └────┬─────┘    └────┬────┘  │
│                      │               │               │        │
│                 findings?       suggestions?     combine     │
│                      │               │               │        │
│                      ▼               ▼               ▼        │
│                 [SECURITY]     [PERFORMANCE]     [DELIVER]    │
└──────────────────────────────────────────────────────────────┘
```

---

## PROCEDIMENTO DETALHADO

### Fase 1: Receive Code

**Input do usuário:**
- `code`: Código fonte para review
- `language`: Linguagem de programação
- `focus`: Foco do review (security / performance / all)
- `severity_threshold`: Mínimo de severidade para reportar

**Log:** `[PHASE:RECEIVE] Code: {lines} lines | Language: {lang}`

### Fase 2: Security Audit

**Ações:**
1. INVOCAR `security-auditor` com o código
2. Coletar vulnerabilidades encontradas

**Log:** `[PHASE:SECURITY] Vulnerabilities: {n} | Risk: {level}`

### Fase 3: Performance Check

**Ações:**
1. INVOCAR `performance-optimizer` com o código
2. Coletar sugestões de otimização

**Log:** `[PHASE:PERFORMANCE] Optimizations: {n} | Score: {score}`

### Fase 4: Unified Report

**Ações:**
1. Combinar findings de segurança e performance
2. Priorizar por severidade
3. Gerar report unificado com score final

**Score Final:**
```
Score = (Security × 0.50) + (Performance × 0.30) + (SOLID Compliance × 0.20)
```

**Log:** `[PHASE:REPORT] Final Score: {score}/100 | Grade: {grade}`

---

## OUTPUT FORMAT

```json
{
  "review_report": {
    "overall_score": 72,
    "grade": "C",
    "summary": "Code has 2 critical security issues and 3 performance opportunities.",
    "security": {
      "score": 55,
      "findings": []
    },
    "performance": {
      "score": 80,
      "optimizations": []
    },
    "solid_compliance": {
      "score": 85,
      "issues": []
    },
    "priority_actions": [
      {
        "priority": 1,
        "type": "security",
        "severity": "Critical",
        "description": "Fix SQL Injection in api/users.py:42"
      }
    ],
    "composition_log": "[full log here]"
  }
}
```

---

## GRADE SCALE

| Score | Grade | Significado |
|-------|-------|-------------|
| 90-100 | A | Excelente - pronto para merge |
| 80-89 | B | Bom - pequenos ajustes |
| 70-79 | C | Adequado - melhorias recomendadas |
| 60-69 | D | Abaixo da média - revisão necessária |
| < 60 | F | Insuficiente - requer refactoring |

---

## CONFIGURAÇÕES

| Parâmetro | Default | Descrição |
|-----------|---------|-----------|
| `severity_threshold` | Low | Mínimo de severidade para reportar |
| `auto_fix` | false | Gerar código corrigido automaticamente |
| `verbose` | false | Logs detalhados |

---

## ERROR HANDLING

### Skill não encontrada
```
SE skill dependente indisponível:
    1. Alertar: "⚠️ [skill] não encontrada"
    2. Oferecer fallback com conhecimento interno
    3. Marcar output como "review parcial"
```

---

## LOGGING ESTRUTURADO

```
[COMPOSE:START] code-review-orchestrator@1.0.0
  [DEP:LOADED] code-standards-base@1.0.0 (summary) [200 tokens]
  [DEP:LOADED] security-auditor@1.0.0 [950 tokens]
  [DEP:LOADED] performance-optimizer@1.0.0 [850 tokens]
[COMPOSE:READY] Total context: 2000 tokens

[PHASE:RECEIVE] Code: 250 lines | Language: Python
[PHASE:SECURITY] Vulnerabilities: 3 | Risk: High
[PHASE:PERFORMANCE] Optimizations: 2 | Score: 80
[PHASE:REPORT] Final Score: 72/100 (C) | Time: 30s
```

## Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

- **Author:** [gugastork](https://github.com/gugastork)
- **Source:** [gugastork/agente-skill-oop](https://github.com/gugastork/agente-skill-oop)
- **License:** MIT

Install and usage instructions live in the source repository linked above.

## Pricing

- **Free** — Free

## Security capabilities

Automated source analysis of v0.1.0 — what this tool can access:

- **Network access:** no
- **Filesystem access:** no
- **Shell / process execution:** no
- **Environment & secrets:** no
- **Dynamic code execution:** no

*"Yes" means the capability is present in the source — more access means more to trust, not that it is unsafe.*


## Versions

- **0.1.0** — security scan: passed — Imported from the upstream source.

## Links

- Listing page: https://agentstack.voostack.com/l/skill-gugastork-agente-skill-oop-code-review-orchestrator
- Seller: https://agentstack.voostack.com/s/gugastork
- Browse the marketplace: https://agentstack.voostack.com/browse

---
Listed on AgentStack — the marketplace for AI agent skills and MCP servers. Every listing is security-reviewed. Creators keep 70%.