Claude Skills by majinmagros

Use when auditing or cleaning up the skill collection — scoring clarity, validating frontmatter, ranking, deduplicating. Triggers on "audita as skills", "dá nota pras skills", "revisa skills", "scorecard de skills", "quais skills estão ruins". Produces a markdown scorecard ranking skills worst to best with one high-value fix per skill.

Use after spec/plan/tasks exist, before implementing, to cross-check artifacts for inconsistencies. Triggers on "/auditoria-artefatos", "audita os artefatos", "consistência entre spec e plano", "spec, plano e tarefas batem?", "análise cruzada dos documentos". Análise READ-ONLY de consistência entre requisitos, plano técnico e lista de tarefas — reporta duplicações, ambiguidades, lacunas de cobertura e violações de constituição com severidade. Não modifica nada.

Use quando precisar auditar papers, datasets, benchmarks e registros científicos com IA — detecção de erros em dados de referência (ex: pontos de ebulição 75 anos), baixa reprodutibilidade (1/3 papers ICML), taxas de erro crescentes (NeurIPS +55% em 4 anos). Triggers em "auditoria científica", "reprodutibilidade papers", "erro dados referência", "SI Labs", "auditoria papers IA", "verificação benchmarks", "dados científicos errados".

Use quando precisar criar agentes que se auto-melhoram continuamente (autobots) — avaliador independente, auto-retrain, ciclo de melhoria autônomo, integração CRM/Slack/GitHub/broker. Triggers em "autobots", "auto-melhoria IA", "auto-retrain", "avaliador independente", "Abacos", "ciclo melhoria autônomo", "IA que aprende sozinha", "auto-improvement loop".

Use when a task is predictable and should become a script instead of AI reasoning — or when deciding what to automate vs keep with the model. Triggers on "automatiza", "script pra isso", "tarefa previsível", "a+b=c", "postar automático", "não precisa pensar", "ganhar eficiência". Automates deterministic steps; keeps creative/reasoning with the LLM.

Evidence-first automation inventory and overlap audit workflow for ECC. Use when the user wants to know which jobs, hooks, connectors, MCP servers, or wrappers are live, broken, redundant, or missing before fixing anything.

Transform Claude Code into a fully autonomous agent system with persistent memory, scheduled operations, computer use, and task queuing. Replaces standalone agent frameworks (Hermes, AutoGPT) by leveraging Claude Code's native crons, dispatch, MCP tools, and memory. Use when the user wants continuous autonomous operation, scheduled tasks, or a self-directing agent loop.

DEPRECATED — use continuous-agent-loop. Legacy patterns for autonomous Claude Code loops (sequential pipelines, REPL, infinite loops, PR loops, de-sloppify, Ralphinho DAG). Triggers on \"autonomous-loops\", \"autonomous loops\", \"loops\".

Build 100% autopilot IG growth factory (Apify top 20% -> filter -> HeyGen+ElevenLabs -> Remotion/FFmpeg -> Metricool) with 3-trigger split. Use when need 0% human after setup, scrape->filter->produce->post. Triggers on \"autopilot content\", \"Apify\", \"HeyGen\", \"Metricool\", \"autopilot IG\"

Use when a new AI tool/service/platform is announced or proposed and you must decide whether to adopt/sign up or keep the current stack — before anyone pays or migrates. Triggers on "vale a pena essa ferramenta?", "avalia essa ferramenta de IA", "assino ou não?", "tool X muda tudo", "devo adotar", "lock-in", "hype", "testar antes de assinar", "build vs buy", "ferramenta nova". Runs a skeptic framework: real scope vs marketing, true cost of the defining feature, lock-in and exit cost, comparis...

Use when backend architecture patterns, API design, database optimization, and server-side best practices for Node.js, Express, and Next.js API routes. Triggers on \"backend-patterns\", \"backend patterns\", \"patterns\".

Use when the user wants to download a song/audio from YouTube. Triggers on "baixar música", "baixar faixa", "baixar som", "baixar do YouTube", "baixar som de <artista>", "quero essa música", "yt download". Given a music name + artist, finds the video on YouTube, confirms with the user, and downloads the audio using baixar_audio.ps1.

"Use when use after competitive-platform-analysis has produced a tiered competitor set. Scores each competitor across nine weighted dimensions (positioning, voice, visual craft, offer packaging, evidence, enterprise-readiness, thought leadership, pricing... Triggers on \"benchmark-methodology\", \"benchmark methodology\", \"methodology\"."license: MIT

Use when the user asks to make something faster, try many variants, run recursive optimization, benchmark latency/throughput/cost, or choose the best implementation by repeated measured tests.

Use when use this skill to measure performance baselines, detect regressions before/after PRs, and compare stack alternatives. Triggers on \"benchmark\".

Use when building AI drug discovery + aging clocks validation pipeline — target ID, molecule generation, clinical trial, 6-clock consensus, UK Biobank 55k. Triggers on "biomedical ai pipeline", "drug discovery ai", "aging clocks validation", "insilico medicine", "rentocertibe", "uk biobank comparison".

Use when an AI agent must operate Blender 3D via the blender-mcp MCP server — setup, scene ops, materials, code execution, and the SVG-first cost-gate workflow. Triggers on \"Blender MCP\", \"blender-mcp\", \"uvx blender-mcp\", \"operate Blender\", \"SVG to mesh\", \"blender-mcp-ops\". Non-triggers: read-only rig/pose inspection (use blender-motion-state-inspection), image-to-Three.js/WebGL code (use img2threejs), Unreal/MetaHuman pipelines. Outcome: a working Blender+MCP loop that validates ...

Use when use this skill when inspecting Blender characters, rigs, poses, animation retargeting, ground contact, facing direction, or model-vs-motion alignment where screenshots alone are not enough. Triggers on \"blender-motion-state-inspection\", \"blender motion state inspection\", \"inspection\".

"Use when turn a one-line objective into a step-by-step construction plan for multi-session, multi-agent engineering projects. Each step has a self-contained context brief so a fresh agent can execute it cold. Includes adversarial review gate, dependency graph, parallel step detection, anti-pattern catalo... Triggers on \"blueprint\"."metadata:

Use when turning books, PDFs, e-books, docs into conversable Agent Skills — on-demand chapter fragments, 50x fewer tokens, cleaning diagram dirt, technical vs text-heavy. Triggers on "transformar livro em skill", "PDF em skill", "book to skill", "livro em skill", "fragmentos de livro", "50x tokens", "virgilio book-to-skill".

Use when a brand needs to discover or articulate its identity through structured multi-session interviews — purpose, positioning, audience, voice, 8 modules, master brandbook. Triggers on "brand discovery", "brand identity interview", "brandbook", "brand positioning", "brand voice".

Build a source-derived writing style profile from real posts, essays, launch notes, docs, or site copy, then reuse that profile across content, outreach, and social workflows. Use when the user wants voice consistency without generic AI writing tropes.

Use when use this skill to automate visual testing and UI interaction verification using browser automation after deploying features. Triggers on \"browser-qa\", \"browser qa\".

Use when bun as runtime, package manager, bundler, and test runner. When to choose Bun vs Node, migration notes, and Vercel support. Triggers on \"bun-runtime\", \"bun runtime\", \"runtime\".

Use when implementing Block Buzz workspace — humans and AI agents as teammates with shared workspace, identity keys, VPS relay, worktrees, per-agent model routing, and free-tier (Kimi K3 via OpenRouter). Triggers on "Buzz", "Block Buzz", "workspace teammates", "human-agent teaming", "Kimi K3 free", "starter agents", "bus.xyz", "WSS relay".

Use when use this skill to monitor and verify a deployed URL after releases — checks HTTP endpoints, SSE streams, static assets, console errors, and performance regressions after deploys, merges, or dependency upgrades. Smoke / canary / post-deploy verification. Triggers on \"canary-watch\", \"canary watch\", \"watch\".

Evolucao do unlazy com gates de qualidade visual + relatorio obrigatorio. Use when precisa de entrega caprichada: site/app com acabamento visual, Kanban drag-and-drop, boids/cardume. Triggers on "capricho", "caprichado", "acabamento visual", "entrega premium", "unlazy melhorado".

Use when managing carriers, negotiating rates, evaluating carrier performance, or building freight strategies. Triggers on "carrier onboarding", "freight RFP", "carrier scorecard", "tender acceptance", "routing guide", "lane rate negotiation", "carrier exit".

Otimizador PF de cartões e pontos a partir de 2-6 faturas. Use quando o usuário pedir melhor cartão, milhas, cashback, anuidade, onde concentrar gasto. Triggers PT: otimizar cartões, pontos e milhas, qual cartão usar, cancelar cartão, analisar faturas. Triggers EN: credit card optimizer, points and miles, best card for spending, annual fee analysis.

Prune model output verbosity (preamble, repeated summary, help offers) to save output tokens — source claims −65/−87%. Use when the user says "resposta direta", "sem enrolação", "cortar verbosidade", "economizar tokens de saída", "modo caveman", or in EN "concise output", "cut verbosity", "no preamble", "direct answer only", "caveman mode", "save output tokens".

Use when reviewing or hardening requirements/specs before implementation. Triggers on "/checklist-requisitos", "checklist de requisitos", "validar qualidade dos requisitos", "requisitos estão bem escritos?", "unit tests for English". Gera checklist que testa a QUALIDADE DOS REQUISITOS (completude, clareza, consistência, mensurabilidade) — não a implementação. O agente gera itens, nunca marca [x].

Use when cisco IOS and IOS-XE review patterns for show commands, config hierarchy, wildcard masks, ACL placement, interface hygiene, and safe change-window verification. Triggers on \"cisco-ios-patterns\", \"cisco ios patterns\", \"patterns\".

Use when persistent per-project memory for Claude Code. Auto-loads project context on session start, tracks sessions with git activity, and writes to native memory. Commands run deterministic Node.js scripts — behavior is consistent across model versions. Triggers on \"ck-cli\", \"ck cli\".

Agente Clareza. Use quando o usuário quiser analisar/sintetizar documentos, contratos ou informações dispersas. Modo telescópio (panorama) ou microscópio (cláusulas/riscos). Triggers em "contrato", "analisar documento", "cláusula", "riscos", "negociar", "assinar", "resumo de documento", "research", "due diligence".

Use when a spec/requirements doc exists but is underspecified, before planning or coding. Triggers on "/clarificar", "especificação vaga", "requisitos incompletos", "falta decisão no spec", "ambiguidade nos requisitos", "clarifica o spec". Detecta ambiguidades e pontos de decisão faltando, faz até 5 perguntas dirigidas (uma por vez, com recomendação justificada) e grava as respostas de volta no documento de requisitos.

Use when setting up or optimizing a Claude account from scratch — email strategy, memory import, model routing, project templates, connectors. Triggers on "otimizar conta claude", "configurar claude do zero", "melhorar setup claude", "email da conta claude", "memory import claude", "model routing claude", "project template claude".

Playbook de governanca de modelos Claude por role, entitlements e effort caps. Use para definir defaults, controlar custo-por-tarefa e padronizar modelo+effort+cache. Triggers PT: qual modelo usar, governanca de modelos, controlar custo Claude, effort cap. Triggers EN: which Claude model, model governance, model defaults, control AI cost.

Use when automating Chrome with Claude — autonomous navigation, ethical scraping, multi-tab research, form filling, approval gates. Triggers on "claude chrome automation", "chrome extension automation", "browser automation claude", "web scraping claude", "chrome extension claude".

Use when choosing and configuring Claude connectors — Gmail/Calendar/Drive/Notion prioritization, MCP vs native, granular permissions, OAuth flow. Triggers on "claude connector strategy", "connector strategy claude", "gmail connector claude", "mcp vs nativo", "connector permissions", "oauth flow claude".

Use when applying Claude Cowork patterns — task execution, file ops, folder processing, recurring tasks, approval gates. Triggers on "claude cowork patterns", "cowork patterns claude", "task execution patterns", "file operations patterns", "folder processing patterns", "recurring tasks claude".

Use when orchestrate multi-agent coding tasks via Claude DevFleet — plan projects, dispatch parallel agents in isolated worktrees, monitor progress, and read structured reports. Triggers on \"claude-devfleet\", \"claude devfleet\", \"devfleet\".

Use when building with Anthropic's Managed Agents SDK — Outcomes, Memory (3-tier), Sandboxing, Briefs, Watchtower. Triggers on "managed agents", "outcomes api", "managed agent memory", "sandboxing", "briefs api", "watchtower agent".

Use when auditing CLAUDE.md length and structure — trims, path-specific rules, HTML comments, prompt-for-next-session. Triggers on "claude md auditor", "claude md trim", "claude md path specific rules", "doctor claude", "/doctor".

Use when routing Claude models by task — Sonet daily, Opus/Fable complex, mid-task swap, per-task cost tracking. Triggers on "claude model router", "roteamento modelos claude", "sonet vs opus", "swap model mid task", "cost tracking per task", "model routing strategy".

Use when scaffolding a typed Claude project — course, client-project, automation, product, research templates with docs, instructions, connectors. Triggers on "claude project template", "project template claude", "project structure claude", "template projeto claude".

Política e checklist da janela rolante de monitoramento Claude. Gatilhos PT: retencao de dados Claude, janela de monitoramento 30 dias, misuse multi-sessao, multi-conta, revisao de seguranca. Triggers EN: Claude data retention policy, rolling monitoring window, multi-session misuse detection, covered models review.

Give Claude vision for video — transcript + keyframes (Cloud Video by Brad Bonano). Use when need Claude to watch YouTube/IG/TikTok/local video and summarize. Triggers on \"claude video\", \"cloud video\", \"transcript video\", \"keyframes\", \"Brad Bonano\"

Use when dictating prompts via voice (Sponcle/Whisper) to Claude — voice interview, transcription cleanup, prompt engineering from audio. Triggers on "voice workflow", "Sponcle", "ditado por voz", "entrevista por voz", "Whisper prompt".

Decisão e transição para Zero Data Retention e Frontier Safeguards. Gatilhos PT: zero retention, ZDR Claude, frontier safeguards, BYOK monitoramento, log no cloud do cliente, transicao faseada ZDR. Triggers EN: Claude zero data retention, enterprise frontier safeguards, customer-held monitoring store, BYOK audit, phased ZDR transition.

Rewrite pre-Claude-5 prompts and skills for Fable 5 / Opus 5 (plus GPT-6 Astra appendix): goal-over-steps, why plus definition of done, answer-vs-action boundaries, strip over-triggers and rituals, global managed-prose line, diff-for-approval rewrite with regex anti-pattern scanner. Use when modernizing prompts, auditing CLAUDE.md or skills, fixing over-verbose or over-acting model behavior. Gatilhos PT: reescrever prompts, modernizar skills, modelo verboso demais, age sem pedir, rituais thin...