# Security

> Router para skills de segurança. Use "security" e eu indico qual skill especializada carregar.

- Skill: `paulohfs/security` (Agent Skill)
- Install (CLI): `npx skillmds@latest add paulohfs/security`
- Raw SKILL.md: https://api.skillmd.com/api/skills/paulohfs/security/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Security
- Author: PauloHFS (https://skillmd.com/u/paulohfs)
- Updated: 2026-09-22
- Page: https://skillmd.com/skills/paulohfs/security

---


# Security Skills Router

Ponto de entrada único para as 6 skills de segurança do repositório. Você invoca `security`; eu mostro a tabela; você chama a skill certa.

| Área | Skill | Quando usar (leading words) |
|------|-------|-----------------------------|
| **API / Backend** | `api-security` | endpoint, REST, controller, middleware auth, JWT, OAuth, rate limit, headers HTTP |
| **Arquitetura** | `architecture-security` | arquitetura, threat modeling, STRIDE, DREAD, DFD, Zero Trust, DevSecOps, governance, CI/CD pipeline |
| **Frontend / UI** | `frontend-security` | componente, formulário, SPA, React, Vue, Angular, CSRF, CSP, SSL/TLS, LGPD UI, XSS |
| **Código Geral (Backend)** | `secure-coding` | validação entrada, injeção SQL, XSS, hash senha, bcrypt/argon2, cookie, header, logging, dependências |
| **LLM Runtime Guards** | `llm-guardrails` | guardrail, prompt injection, PII output scan, rate limit LLM, circuit breaker, LGPD compliance, encoding bypass |
| **LLM Risk Assessment** | `llm-security-assessment` | assessment, OWASP LLM 2025, matriz risco, gap analysis, priorização, relatório executivo |

## Como instalar a skill desejada

```bash
npx skills add github:PauloHFS/ai-agent-skills/skills/<skill-name>
```

Exemplos:
```bash
npx skills add github:PauloHFS/ai-agent-skills/skills/api-security
npx skills add github:PauloHFS/ai-agent-skills/skills/llm-guardrails
```

## Decisão rápida

- **Escrevendo código de API?** → `api-security`
- **Desenhando/revisando arquitetura?** → `architecture-security`
- **Escrevendo componente/frontend?** → `frontend-security`
- **Escrevendo qualquer backend?** → `secure-coding`
- **Protegendo LLM em produção?** → `llm-guardrails`
- **Avaliando risco de sistema LLM?** → `llm-security-assessment`