Top Agent Skills

Escreve roteiros de vídeo e VSL com gancho, abertura, desenvolvimento, prova, CTA e fechamento, incluindo marcação de tempo e indicação visual.

Escreve microcopy de interface — mensagens de erro, estados vazios, tooltips, rótulos de botão, onboarding e notificações, com foco em clareza e acessibilidade.

Revisa e melhora copy existente através de 7 passes focados (clareza, voz, e daí, prove, especificidade, emoção, risco zero) somados à detecção determinística de padrões de escrita de IA. Use quando o usuário disser "revisar copy", "melhorar esse texto", "revisão", "está parecendo IA", "polir", "dar uma olhada nesse.

Guides backup, restore, snapshot, PITR, retention, and disaster recovery planning for PostgreSQL and AWS, emphasizing tested restores and RTO/RPO alignment.

Coleta a linguagem literal do público em avaliações, fóruns, redes sociais, tickets de suporte e transcrições de vendas para gerar um banco de frases verbatim que orienta a escrita de copy.

Gerencia issues no YouTrack Cloud via MCP oficial, incluindo busca, criação, comentários, tags, artigos e registro de trabalho, com fallback REST.

Interpreta métricas de copy e aponta o que fazer — conversão de landing, abertura e clique de e-mail, CTR e CPA de anúncio, engajamento de social. Use quando o usuário disser "analisar desempenho", "como foi a campanha", "os números do mês", "essa página não converte", "taxa de abertura caiu", "relatório de.

Implementa e valida dados estruturados schema.org em JSON-LD — Organization, Article, FAQPage, Product, BreadcrumbList. Use quando o usuário disser "schema", "dados estruturados", "JSON-LD", "rich snippet", "FAQ schema", "schema.org".

Documents, standardizes, and governs a Vue 3 + shadcn-vue design system, including component documentation templates, patterns for forms, empty states, loading states, and feedback, plus a consistency checklist and conventional commit conventions.

Projeta métricas, logs, traces, dashboards, alertas e SLO com Prometheus, Grafana, Loki, Tempo, Mimir, Zabbix e Alertmanager, incluindo consultas PromQL e boas práticas de alertas.

Mantém a documentação técnica viva no Notion e no repositório, gerando runbooks, inventários, ADRs e procedimentos a partir da infraestrutura real.

Audits Vue 3 components for design-system consistency, legacy patterns, and missing states, producing a component log.

Diagnoses Linux server issues across CPU, memory, disk, I/O, and network, and manages systemd services, SSH, and Fail2Ban on Ubuntu, Debian, and AlmaLinux.

Escreve texto alternativo acessível para peças de redes sociais, transmitindo a informação da imagem em vez de descrevê-la, com regras por tipo e formato de saída.

Produz vídeo curto com roteiro, storyboard, prompts por cena e geração via Runway quando há chave, ou especificação manual para Higgsfield.

Enforces a canonical UI/UX ruleset with 14 sections and 70+ prioritized rules (P0–P3) for implementing, reviewing, or auditing UI components, referencing tokens from a theme reference.

Interage com o YouTrack via REST API para operações que o MCP oficial não cobre, como boards, sprints, relatórios de tempo, anexos e comandos em lote. Requer YOUTRACK_BASE_URL e token.

Arquiva briefs, peças e aprendizados de campanhas no Obsidian (pasta local) e no Notion (via MCP), usando links e estrutura de pastas para reuso futuro.

Conduz investigação de incidentes com método: triagem, hipóteses, mitigação, RCA e postmortem, priorizando evidências e confirmação humana.

Cria peças únicas de feed (quote card, card de dado, capa, anúncio estático) com hierarquia visual e boas práticas de design, integrando-se ao Figma para montagem e screenshot.

Estrutura prompts para geradores de imagem e vídeo, otimizando custo e consistência com a marca.

Benchmark, load test, capacity plan, and cache with k6, JMeter, Locust, and pgbench. Use when the user says "slow", "performance", "load test", "stress test", "how many users can it handle", "capacity", "cache", "benchmark", "k6".

Avalia a confiabilidade de uma plataforma inteira, mapeando pontos únicos de falha, calculando error budgets a partir de SLOs, revisando prontidão para produção e analisando resiliência e capacidade.

Organiza, compara e seleciona variações de imagens geradas, publicando uma grade de comparação como Artifact para aprovação do time, com critérios de avaliação e registro de decisões.

Cria e mantém o contexto de criativos da marca, incluindo Figma, brand kit, formatos, geradores e restrições, para uso por outras skills do plugin.

Audita criativos antes da aprovação, verificando contraste, legibilidade, área segura, hierarquia, aderência ao brand kit e conformidade, com veredito de aprovação, ajuste ou reprovação.

Monta carrosséis completos para Instagram, LinkedIn e Facebook, estruturando slides, definindo hierarquia e construindo no Figma como componente com variantes, com revisão e texto alternativo.

Generates a structured Product Requirements Document (PRD) for a new feature, including clarifying questions, user stories, functional requirements, and success metrics.

Converts existing PRDs into the prd.json format used by the Ralph autonomous agent system, including splitting large features into right-sized stories and creating a progress file.

Decompose large user stories into thin, independently deliverable vertical slices using the SPIDR method (Spike, Path, Interface, Data, Rules).

Reviews a codebase for structural smells and refactors them with provable safety, using micro commits and green-bar discipline.

Apply strict-mode Bash scripting conventions: minimal validation, concise logic, executable scripts, and no unnecessary comments or echoes.

Clarifies ambiguous or conflicting requests by researching first, then asking only judgment calls. Use when prompts say "grill me", ask hard questions, request relentless interrogation, pressure-test assumptions, clarify scope/requirements, define success criteria, or request system-design/optimization decisions.

Compiles up to five durable insights from the current conversation for possible addition to CLAUDE.md, focusing on patterns, preferences, and project knowledge.

Writes user-focused acceptance criteria for Jira stories using Given/When/Then scenarios, and updates the ticket via the Atlassian MCP server.

Mines a codebase for evidence-backed improvement opportunities, forcing two escalation gates to surface breakthrough ideas, and outputs a ranked portfolio with a plan seed without implementing.

Audits over-engineered codebases by factoring layers into live obligations, quotienting redundant distinctions, ablating unearned surface, and normalizing survivors while preserving required behavior. Produces an evidence-backed Reduction Certificate, cuts, migration phases, proof signals, and rollback.

Runs a ten-round adversarial gauntlet to pressure-test absolute claims, refining them with explicit boundaries and an Oracle synthesis.

Acceptance-test newly added features using browser automation, happy-path flows, screenshots, UX notes, and a written report.

Refactors code by renaming variables, methods, classes, and types to improve clarity and intent, with automatic testing and commits after each rename.

Classifies code review findings to choose the smallest correct response, preventing one-patch-per-comment churn.

Define owned inductive invariants and enforce them at parse, construct, API, DB, lock, or transaction boundaries, with counterexample-driven verification.

Performs comprehensive exploratory testing on a website using browser automation, covering user journeys, forms, errors, accessibility, responsiveness, performance, and security basics.

Verifies that tests for already-implemented behavior are meaningful by temporarily breaking the production code and confirming the test fails for the expected reason.

Systematically explore an unfamiliar codebase and build a reusable architecture summary with file:line evidence, starting from documentation and tracing data flow.

Rereads code, specs, plans, reviews, or skill edits with fresh eyes to catch blunders, oversights, and omissions before closure, reporting concrete evidence-backed findings.

Provides guidance on Hong Kong's four mobile network operators for reference and selection.

Operates AWS resources directly through the AWS CLI, covering S3, EC2, Lambda, and CloudFormation. Activates when AWS-related terms are mentioned and enforces confirmation for write operations.

Operates Google Cloud Platform resources via the gcloud CLI, covering compute instances, storage buckets, BigQuery, Cloud Run, GKE, IAM, and billing. It checks the environment, authenticates, confirms destructive operations, and diagnoses common errors.

Creates standardized git commits using the Conventional Commits specification by analyzing the diff, staging files, and generating appropriate commit messages.

Operates Qiniu Cloud object storage and CDN through the qshell CLI, covering bucket listing, file upload/download, batch operations, and CDN refresh.

Operate Upyun cloud storage via the upx CLI, covering file upload, download, deletion, directory management, sync, and CDN purge.

Guides users through opening Hong Kong bank accounts with ZA Bank and HSBC HK.

Performs web search, content extraction, site crawling, URL discovery, and deep research via the Tavily CLI, returning JSON for LLM consumption.

Discovers and lists all URLs on a website without extracting content, using the Tavily CLI. Faster than crawling, with options for depth, filtering, and natural language instructions.

Provides a template for creating agent skills with usage guidance and step-by-step instructions.

Generates comprehension questions and language exercises based on analyzed content to create a complete learning loop.

Operates Hetzner Cloud resources via the hcloud CLI, covering server, image, network, and load balancer management with read-only and write operations.

Crawls websites and saves content from multiple pages as local markdown files using the Tavily CLI, with options for depth, breadth, path filtering, and semantic focus.

Reviews all uncommitted changes in the current Git repository and produces a structured code review report with findings and a verdict.