observability
Skill1 starsChanged 4 months agoMITView on GitHub
Observabilidade production-grade: logs, métricas e traces (os 3 pilares), SLI/SLO/error budget, alertas acionáveis, cardinalidade e OpenTelemetry. Use quando o usuário mencionar: "observabilidade", "monitoramento", "logs", "métricas", "tracing", "SLO", "SLI", "alerta", "dashboard de sistema", "por que está lento em prod", "APM", "OpenTelemetry", "Prometheus".
87 other projects use a version of this.
Adopted by
Most recently taken up by
- juninmd/skillstheir own version, first seen Nov 8, 2026
- domengabrovsek/agent-configtheir own version, first seen Nov 8, 2026
- domengabrovsek/claudetheir own version, first seen Nov 8, 2026
- Ingvord/legendary-bassoontheir own version, first seen Nov 8, 2026
- sundaram2021/claude-code-observability-systemtheir own version, first seen Nov 8, 2026
- Mehul72/skillstheir own version, first seen Nov 8, 2026
Used with
Projects with it also use gitops, background-jobs and 8 more.
Other files people use alongside this one.
Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
kcns008Skillgitops
GitOps Agent (Flow) — manages ArgoCD applications, Helm charts, Kustomize overlays, deployment strategies (canary, blue-green, rolling), multi-cluster GitOps, and drift detection for Kubernetes and OpenShift clusters.
Also in 4 projects that use this
YepAPISkillbackground-jobs
Queue workers with BullMQ/Inngest/Trigger.dev, job retries, dead letter queues, concurrency.
Also in 4 projects that use this
Mushfiqur-Rahman-RobinSkillrag-systems
Best practices for building Retrieval-Augmented Generation systems — chunking, embedding, hybrid retrieval, reranking, indexing, caching, evaluation, observability — plus a catalog of RAG patterns and when each applies.
Also in 3 projects that use this
YepAPISkilldata-pipelines
ETL patterns, cron scheduling, queue-based processing, and batch jobs with Inngest/BullMQ.
Also in 4 projects that use this
wpankSkillcaching
Caching strategies, invalidation, eviction policies, HTTP caching, distributed caching, and anti-patterns. Use when designing cache layers, choosing eviction policies, debugging stale data, or optimizing read-heavy workloads.
Also in 5 projects that use this
MicrosoftSkilldata-modeling
Use when the app needs to store or read data — records, a database, entities, CRUD, lists, "save/remember X", or per-user data with row-level security. This app's data layer is Rayfin's: you declare entities as TypeScript classes with decorators in rayfin/data/, register them in the schema, and read/write them through the typed rayfin-client. Covers: defining an @entity, field decorators, registering it, per-entity access with @role, owner-only row-level security, and the client query/mutation API. Triggers: data, database, model, entity, schema, table, record, CRUD, create, read, update, delete, list, store, save, persist, per-user, row-level security, RLS, access control, ownership.
Also in 6 projects that use this
Other versions87 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 87
- openai/plugins★ 7.8ksame name
- BuilderIO/agent-native★ 7.6ksame name
- brianlovin/briOS★ 2.3ksame name
- lablup/backend.ai★ 679same name
- DevelopersGlobal/ai-agent-skills★ 216same name
- kid-sid/claude-spellbook★ 190same name
- YougLin-dev/Aha-Loop★ 181same name
- proyecto26/system-design-skills★ 118same name
- TheBeardedBearSAS/claude-craft★ 108same name
- AsiaOstrich/universal-dev-standards★ 84same name
77 more not listed
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository70 other files it gives its agents.
- agent-creationSkill
- handoffSkill
- task-distributorSkill
- task-plannerSkill
- accessibilitySkill
- airbnb-designSkill
- airtable-designSkill
- atlassian-designSkill
- carbon-designSkill
- cloudscape-designSkill
- coinbase-designSkill
- discord-designSkill
- duolingo-designSkill
- figma-designSkill
- fluent-designSkill
- linear-designSkill
- material-designSkill
- meta-designSkill
- netflix-designSkill
- notion-designSkill
- nubank-designSkill
- one-ui-designSkill
- pinterest-designSkill
- polaris-designSkill
- premium-web-designSkill
- primer-designSkill
- reddit-designSkill
- revolut-designSkill
- robinhood-designSkill
- salesforce-lightning-designSkill
- slack-designSkill
- spectrum-designSkill
- spotify-designSkill
- stripe-designSkill
- tiktok-designSkill
- twitch-designSkill
- uber-base-designSkill
- vercel-geist-designSkill
- visionos-designSkill
- whatsapp-designSkill
- economicsSkill
- legal-basicsSkill
- negotiationSkill
- politicalSkill
- api-designSkill
- architectureSkill
- caching-strategiesSkill
- ci-cdSkill
- database-designSkill
- event-drivenSkill
- payment-systemsSkill
- performanceSkill
- qaSkill
- refactoringSkill
- securitySkill
- system-designSkill
- testingSkill
- clone-websiteSkill
- graphifySkill
- remotionSkill
- The first 60 shown.
The file
Files
Observability
Você atua como engenheiro de confiabilidade. Objetivo: instrumentar sistemas para responder perguntas novas sem redeploy — não só "está no ar?", mas "por que este usuário viu latência às 3h12?". Observabilidade é sobre conseguir explicar o desconhecido.
Os 3 pilares (use os três, cada um resolve algo)
| Pilar | Responde | Cuidado |
| Logs | O que aconteceu neste evento (detalhe) | Estruturados (JSON), não texto solto; sem PII/segredo; custo de volume |
| Métricas | Quantos/quão rápido, ao longo do tempo (agregado) | Baratas e rápidas, mas baixa cardinalidade |
| Traces | O caminho de UMA requisição por N serviços | Contexto propagado (trace-id); amostragem |
Regra: métrica te alerta → trace te localiza → log te explica.
Logs que prestam
- Estruturados: JSON com campos (
level,trace_id,user_id,latency_ms), não string interpolada. - Correlacionáveis: todo log carrega
trace_id/request_idpara casar com traces. - Níveis com disciplina: ERROR = ação humana necessária; WARN = anômalo mas tratado; INFO = marco; DEBUG = fora de prod.
- Sem vazar: nunca logue senha, token, PAN, PII sensível.
Métricas — os 4 sinais dourados (Google SRE)
Instrumente todo serviço com:
- Latency — p50/p95/p99 (média mente; use percentis).
- Traffic — req/s, throughput.
- Errors — taxa de falha (5xx, exceptions).
- Saturation — quão cheio (CPU, memória, fila, conexões).
Cardinalidade: cada combinação única de labels é uma série temporal.
user_idcomo label de métrica explode custo — use em log/trace, não em métrica.
Traces (distribuído)
- Propague o contexto (trace-id, span-id) entre serviços via headers (W3C Trace Context).
- Cada span mede uma operação; a árvore mostra onde o tempo foi.
- Amostre (head/tail sampling) para controlar custo sem perder os traces interessantes (erros/lentos).
SLI / SLO / Error budget (o que realmente importa)
- SLI — indicador medível da experiência (ex: % de requests < 300ms e 2xx).
- SLO — a meta (ex: 99,9% num mês).
- Error budget — o complemento (0,1% ≈ 43 min/mês). Enquanto há budget, entregue features; se estourou, pare e conserte confiabilidade.
- Alerte sobre sintoma, não causa — alerte que o SLO está queimando (usuário sofrendo), não que "CPU 80%" (pode ser inofensivo).
Alertas acionáveis (senão viram ruído)
Todo alerta deve: (a) indicar impacto real ao usuário, (b) ser acionável agora, (c) apontar o runbook. Alerta que ninguém age → delete. Fadiga de alerta mata a resposta a incidentes.
Fluxo de instrumentação
- Defina o SLI/SLO do serviço (o que é "bom" para o usuário).
- Instrumente os 4 sinais dourados + traces nos caminhos críticos.
- Logs estruturados com trace-id.
- Dashboards: visão de saúde (golden signals) + drill-down.
- Alertas sobre queima de error budget, com runbook ([[runbook]] / [[handoff]] para on-call).
Recursos externos
| Recurso | O que pegar |
| Google SRE Book | SLI/SLO/error budget, golden signals, alerting. |
| OpenTelemetry | Padrão vendor-neutral de traces/métricas/logs. |
| Prometheus | Métricas + PromQL + alerting. |
| Honeycomb — Observability | Observabilidade orientada a eventos e alta cardinalidade. |
| Grafana | Dashboards e visualização dos 3 pilares. |
| W3C Trace Context | Propagação padronizada de trace-id. |
Do / Don't
| Do | Don't |
| Logs estruturados + trace-id | print/string interpolada |
| Percentis (p95/p99) | Só média de latência |
| Alertar por sintoma/SLO | Alertar por "CPU alta" |
| Baixa cardinalidade em métrica | user_id como label de métrica |
Checklist
- SLI/SLO definidos por serviço
- 4 sinais dourados instrumentados (latency p95/p99, traffic, errors, saturation)
- Traces com contexto propagado nos caminhos críticos
- Logs estruturados, com trace-id, sem PII/segredo
- Alertas acionáveis sobre queima de error budget + runbook
- Cardinalidade de métricas controlada
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

