AGENT SKILL

skill-sentinel

Auditoria e evolucao do ecossistema de skills. Qualidade de codigo, seguranca, custos, gaps, duplicacoes, dependencias e relatorios de saude.

License 许可
MIT
最近更新
2026-11-10
Stars
★ 1
Forks
0

01 前言 Front Matter

name
skill-sentinel
risk
safe
tags
[ "governance", "audit", "quality", "skill-health" ]
tools
[ "claude-code", "antigravity", "cursor", "gemini-cli", "codex-cli" ]
author
renat
source
community
date_added
2026-03-06
description
Auditoria e evolucao do ecossistema de skills. Qualidade de codigo, seguranca, custos, gaps, duplicacoes, dependencias e relatorios de saude.

02 资源文件 21

03 SKILL.md 正文

Skill Sentinel

Overview

Auditoria e evolucao do ecossistema de skills. Qualidade de codigo, seguranca, custos, gaps, duplicacoes, dependencias e relatorios de saude.

When to Use This Skill

  • When the user mentions "auditar skills" or related topics
  • When the user mentions "qualidade skills" or related topics
  • When the user mentions "verificar skills ecossistema" or related topics
  • When the user mentions "saude ecossistema skills" or related topics
  • When the user mentions "skills duplicadas" or related topics
  • When the user mentions "otimizar skills" or related topics

Do Not Use This Skill When

  • The task is unrelated to skill sentinel
  • A simpler, more specific tool can handle the request
  • The user needs general-purpose assistance without domain expertise

How It Works

Meta-agente que monitora, audita e evolui o ecossistema de skills. Analisa todas as skills em 7 dimensoes, identifica problemas, sugere melhorias e recomenda novas skills especialistas.

Resumo Rapido

Area Script O que faz
Discovery scanner.py Descobre todas as skills automaticamente
Qualidade analyzers/code_quality.py Complexidade, docstrings, error handling
Seguranca analyzers/security.py Secrets, SQL injection, HTTPS
Performance analyzers/performance.py API calls, caching, retry
Governanca analyzers/governance_audit.py Rate limits, audit log, confirmacoes
Documentacao analyzers/documentation.py SKILL.md, triggers, references
Dependencias analyzers/dependencies.py requirements.txt, versoes
Cross-Skill analyzers/cross_skill.py Duplicacao, padroes compartilhados
Custos cost_optimizer.py Tokens, verbosidade, output
Recomendacoes recommender.py Gap analysis, novas skills
Relatorio report_generator.py Markdown estruturado
Orquestracao run_audit.py CLI principal

Localizacao

C:\Users\renat\skills\skill-sentinel\
├── SKILL.md
├── scripts/
│   ├── requirements.txt
│   ├── config.py
│   ├── db.py
│   ├── governance.py
│   ├── scanner.py
│   ├── analyzers/
│   │   ├── code_quality.py
│   │   ├── security.py
│   │   ├── performance.py
│   │   ├── governance_audit.py
│   │   ├── documentation.py
│   │   ├── dependencies.py
│   │   └── cross_skill.py
│   ├── recommender.py
│   ├── cost_optimizer.py
│   ├── report_generator.py
│   └── run_audit.py
├── references/
│   ├── analysis_criteria.md
│   ├── security_patterns.md
│   ├── skill_template.md
│   └── schema.md
└── data/
    ├── sentinel.db
    └── reports/

Instalacao

pip install -r C:\Users\renat\skills\skill-sentinel\scripts\requirements.txt

Comandos Principais


## Auditoria Completa De Todas As Skills

python C:\Users\renat\skills\skill-sentinel\scripts\run_audit.py

## Auditar Apenas Uma Skill

python C:\Users\renat\skills\skill-sentinel\scripts\run_audit.py --skill instagram

## Apenas Recomendacoes De Novas Skills

python C:\Users\renat\skills\skill-sentinel\scripts\run_audit.py --recommend

## Comparar Com Auditoria Anterior (Tendencias)

python C:\Users\renat\skills\skill-sentinel\scripts\run_audit.py --compare

## Output Em Json (Para Processamento)

python C:\Users\renat\skills\skill-sentinel\scripts\run_audit.py --format json

## Ver Historico De Auditorias

python C:\Users\renat\skills\skill-sentinel\scripts\run_audit.py --history

## Descobrir Skills Disponiveis

python C:\Users\renat\skills\skill-sentinel\scripts\scanner.py

## Ver Audit Log Do Sentinel

python C:\Users\renat\skills\skill-sentinel\scripts\governance.py

## Verificar Banco De Dados

python C:\Users\renat\skills\skill-sentinel\scripts\db.py

1. Qualidade De Codigo (Peso: 20%)

  • Complexidade ciclomatica por funcao (limiar: 10)
  • Tamanho de funcoes (limiar: 50 linhas)
  • Tamanho de arquivos (limiar: 500 linhas)
  • Cobertura de docstrings
  • Padroes de error handling (bare except, broad except)

2. Seguranca (Peso: 20%)

  • Secrets hardcoded (tokens, passwords, API keys)
  • SQL injection (f-strings em queries)
  • URLs HTTP inseguras
  • Tokens em logs
  • Validacao de input

3. Performance (Peso: 15%)

  • Retry com backoff para APIs
  • Timeouts configurados
  • Reuso de conexoes HTTP
  • N+1 queries
  • Async/concorrencia

4. Governanca (Peso: 15%)

  • Nivel 0: Nenhuma
  • Nivel 1: Action logging
  • Nivel 2: Logging + rate limiting
  • Nivel 3: Completa (+ confirmacoes 2-step)
  • Nivel 4: Avancada (+ alertas e trends)

5. Documentacao (Peso: 15%)

  • SKILL.md com frontmatter (name, description, version)
  • Trigger keywords (PT-BR e EN)
  • Secoes obrigatorias e recomendadas
  • Reference files

6. Dependencias (Peso: 15%)

  • requirements.txt presente
  • Versoes pinadas
  • Deps importadas vs listadas
  • Deps listadas vs importadas

7. Cross-Skill (Analise Global)

  • Modulos duplicados entre skills
  • Padroes de Database compartilhados
  • Governanca inconsistente
  • Oportunidades de extracao

Otimizacao De Custos

Alem das 7 dimensoes, o sentinel analisa impacto de custo:

  • Tamanho do SKILL.md (tokens consumidos por ativacao)
  • References grandes sem indice
  • Output verboso dos scripts
  • Ausencia de output JSON estruturado

Gap Analysis E Recomendacoes

O recommender identifica capacidades ausentes no ecossistema comparando com uma taxonomia de 20 categorias e gera templates de SKILL.md prontos para novas skills sugeridas.

Governanca Do Sentinel

O proprio sentinel pratica o que prega:

  • Todas as auditorias sao registradas em action_log
  • Historico de scores em score_history para tendencias
  • Relatorios salvos em data/reports/

Workflows Comuns

1. Primeira auditoria do ecossistema:

python run_audit.py

Gera relatorio completo com scores, findings e recomendacoes.

2. Monitorar evolucao ao longo do tempo:

python run_audit.py --compare

Mostra delta de scores entre auditorias.

3. Validar uma skill antes de deploy:

python run_audit.py --skill nome-da-skill

Auditoria focada com findings especificos.

4. Identificar proxima skill a criar:

python run_audit.py --recommend

Gap analysis com templates prontos.

Formato Do Relatorio

O relatorio gerado em data/reports/ contem:

  1. Resumo executivo (tabela de scores)
  2. Tendencias (se houver auditoria anterior)
  3. Findings por severidade (critico/alto/medio/baixo/info)
  4. Analise por skill (detalhada)
  5. Recomendacoes de novas skills
  6. Plano de acao priorizado

Referencias

Para detalhes tecnicos, consultar:

  • references/analysis_criteria.md - Rubricas de scoring
  • references/security_patterns.md - Padroes de seguranca
  • references/skill_template.md - Template para novas skills
  • references/schema.md - Schema do banco de dados

Best Practices

  • Provide clear, specific context about your project and requirements
  • Review all suggestions before applying them to production code
  • Combine with other complementary skills for comprehensive analysis

Common Pitfalls

  • Using this skill for tasks outside its domain expertise
  • Applying recommendations without understanding your specific context
  • Not providing enough project context for accurate analysis

Related Skills

  • skill-installer - Complementary skill for enhanced analysis

Limitations

  • Use this skill only when the task clearly matches the scope described above.
  • Do not treat the output as a substitute for environment-specific validation, testing, or expert review.
  • Stop and ask for clarification if required inputs, permissions, safety boundaries, or success criteria are missing.

04 同仓库更多 Skills

YYC-Cube
YYC-Cube/YYC3-Knowledge-Base

vercel-sandbox

Run agent-browser + Chrome inside Vercel Sandbox microVMs for browser automation from any Vercel-deployed app. Use when the user needs browser automation in a Vercel app (Next.js, SvelteKit, Nuxt, Remix, Astro, etc.), wants to run headless Chrome without binary size limits, needs persistent browser sessions across commands, or wants ephemeral isolated browser environments. Triggers include "Vercel Sandbox browser", "microVM Chrome", "agent-browser in sandbox", "browser automation on Vercel", or any task requiring Chrome in a Vercel Sandbox.

★ 1
YYC-Cube
YYC-Cube/YYC3-Knowledge-Base

slack

Interact with Slack workspaces using browser automation. Use when the user needs to check unread channels, navigate Slack, send messages, extract data, find information, search conversations, or automate any Slack task. Triggers include "check my Slack", "what channels have unreads", "send a message to", "search Slack for", "extract from Slack", "find who said", or any task requiring programmatic Slack interaction.

★ 1
YYC-Cube
YYC-Cube/YYC3-Knowledge-Base

electron

Automate Electron desktop apps (VS Code, Slack, Discord, Figma, Notion, Spotify, etc.) using agent-browser via Chrome DevTools Protocol. Use when the user needs to interact with an Electron app, automate a desktop app, connect to a running app, control a native app, or test an Electron application. Triggers include "automate Slack app", "control VS Code", "interact with Discord app", "test this Electron app", "connect to desktop app", or any task requiring automation of a native Electron application.

★ 1
YYC-Cube
YYC-Cube/YYC3-Knowledge-Base

dogfood

Systematically explore and test a web application to find bugs, UX issues, and other problems. Use when asked to "dogfood", "QA", "exploratory test", "find issues", "bug hunt", "test this app/site/platform", or review the quality of a web application. Produces a structured report with full reproduction evidence -- step-by-step screenshots, repro videos, and detailed repro steps for every issue -- so findings can be handed directly to the responsible teams.

★ 1
YYC-Cube
YYC-Cube/YYC3-Knowledge-Base

agent-browser

Browser automation CLI for AI agents. Use when the user needs to interact with websites, including navigating pages, filling forms, clicking buttons, taking screenshots, extracting data, testing web apps, or automating any browser task. Triggers include requests to "open a website", "fill out a form", "click a button", "take a screenshot", "scrape data from a page", "test this web app", "login to a site", "automate browser actions", or any task requiring programmatic web interaction.

★ 1
YYC-Cube
YYC-Cube/YYC3-Knowledge-Base

shadcn

Manages shadcn components and projects — adding, searching, fixing, debugging, styling, and composing UI, including chat interfaces. Provides project context, component docs, and usage examples. Applies when working with shadcn/ui, component registries, presets, --preset codes, or any project with a components.json file. Also triggers for "shadcn init", "create an app with --preset", "switch to --preset", and for building a design system or applying a DESIGN.md or brand spec to shadcn/ui.

★ 1