Skill: ops-monitoring

Inline

Application instrumentation for monitoring. Trigger when the user wants to add logs, metrics, or traces.

Configuration

Property	Value
Context	inline
Pre-approved tools	None declared: the project's permission settings apply
Trigger	manual only — run /ops-monitoring

Detailed description

Monitoring Instrumentation

3 Pillars of Observability

1. Logs - Discrete events
2. Metrics - Numerical measurements
3. Traces - Request paths

Structured Logs (Node.js)

import pino from 'pino';

const logger = pino({
  level: process.env.LOG_LEVEL || 'info',
  base: { service: 'api', env: process.env.NODE_ENV },
});

logger.info({ userId: '123', action: 'login' }, 'User logged in');
logger.error({ err, requestId }, 'Request failed');

Prometheus Metrics

import { Counter, Histogram, Registry } from 'prom-client';

const httpRequests = new Counter({
  name: 'http_requests_total',
  help: 'Total HTTP requests',
  labelNames: ['method', 'path', 'status'],
});

const httpDuration = new Histogram({
  name: 'http_request_duration_seconds',
  help: 'Request duration',
  labelNames: ['method', 'path'],
  buckets: [0.1, 0.5, 1, 2, 5],
});

OpenTelemetry Traces

import { trace } from '@opentelemetry/api';

const tracer = trace.getTracer('my-service');

async function processOrder(orderId: string) {
  return tracer.startActiveSpan('processOrder', async (span) => {
    span.setAttribute('orderId', orderId);
    try {
      // ... processing
    } finally {
      span.end();
    }
  });
}

Health Checks

app.get('/health', (req, res) => res.json({ status: 'ok' }));

app.get('/ready', async (req, res) => {
  const dbOk = await db.query('SELECT 1');
  res.status(dbOk ? 200: 503).json({ db: dbOk });
});

Deploying the observability stack

Once the code is instrumented, deploy the backing stack (Prometheus + Grafana + Loki + Alertmanager).

• Mode: Docker Compose for dev/staging, Kubernetes + Helm for production (or Victoria Metrics / managed).
• Prometheus: prometheus.yml scrape configs + alert.rules.yml.
• Alertmanager: alertmanager.yml routes + receivers (Slack / PagerDuty / email).
• Grafana: provision datasources + dashboards.
• Loki + Promtail: log aggregation; add node-exporter + cAdvisor for system metrics.
• Persistent storage for metrics data; never expose Prometheus/Alertmanager without auth in production; configure alerts before going to prod and test the stack in staging first.

See also

Grafana Labs publishes their own official agent skills at grafana/skills (31★, last commit 2026-05-04). The repo covers Grafana Core, Grafana Cloud, the LGTM stack (Loki/Grafana/Tempo/Mimir), k6 performance testing, and the Grafana app SDK. A separate companion repo grafana/pyroscope-skills covers continuous profiling.

When working on a project that uses the Grafana / LGTM stack, install the vendor skill alongside this one. This skill captures the three-pillar instrumentation overview (logs / metrics / traces) and the foundation's basic OTEL + health-check skeleton; the vendor skill captures the canonical Grafana operational patterns that evolve with each Grafana release. For non-Grafana stacks (Datadog, New Relic, Honeycomb, etc.), this skill remains the primary reference.

Vendor-neutrality: Grafana Labs is independent. No concern.

Install command and full list of validated vendor skills: docs/recipes/recommended-vendor-skills.md. Audit pilot trace: specs/marketplace-audit/ops-skills-pilot-2026-05-06.md.

Manual invocation only

This skill carries disable-model-invocation: true, so Claude cannot load it — not when keywords match, and not from inside the same-named command either. Start it yourself with /ops-monitoring.

Context inline

Inline means the skill runs in the conversation (the default, no context key):

• Sees the full history
• Its work is visible as it happens
• Right for almost every skill

---

Practical examples

1. Example: Structured Logging + Prometheus Metrics

Example: Structured Logging + Prometheus Metrics

Scenario

A Node.js API needs structured JSON logging for observability and Prometheus metrics for alerting.

Structured Logging with Pino

// src/lib/logger.ts
import pino from 'pino';

export const logger = pino({
  level: process.env.LOG_LEVEL || 'info',
  formatters: {
    level: (label) => ({ level: label }),
  },
  redact: ['req.headers.authorization', 'password', 'token'],
  serializers: {
    err: pino.stdSerializers.err,
    req: pino.stdSerializers.req,
  },
});

// Usage in route handler
export function createOrder(req, res) {
  const log = logger.child({ requestId: req.id, userId: req.user.id });
  log.info({ orderId: order.id, amount: order.total }, 'order created');
  // Output: {"level":"info","requestId":"abc","userId":"u1","orderId":"o1","amount":99.99,"msg":"order created"}
}

Prometheus Metrics

// src/lib/metrics.ts
import { Counter, Histogram, Registry } from 'prom-client';

export const registry = new Registry();

export const httpRequestDuration = new Histogram({
  name: 'http_request_duration_seconds',
  help: 'Duration of HTTP requests in seconds',
  labelNames: ['method', 'route', 'status_code'],
  buckets: [0.01, 0.05, 0.1, 0.5, 1, 5],
  registers: [registry],
});

export const httpRequestTotal = new Counter({
  name: 'http_requests_total',
  help: 'Total number of HTTP requests',
  labelNames: ['method', 'route', 'status_code'],
  registers: [registry],
});

// Middleware
export function metricsMiddleware(req, res, next) {
  const end = httpRequestDuration.startTimer();
  res.on('finish', () => {
    const labels = { method: req.method, route: req.route?.path || req.path, status_code: res.statusCode };
    end(labels);
    httpRequestTotal.inc(labels);
  });
  next();
}

Metrics Endpoint

// GET /metrics
app.get('/metrics', async (req, res) => {
  res.set('Content-Type', registry.contentType);
  res.end(await registry.metrics());
});

Key Decisions

• Pino over Winston: 5x faster, native JSON output, lower memory
• Redact sensitive fields: Authorization headers and passwords auto-masked
• Child loggers: Add requestId context without passing it everywhere
• Histogram buckets: Tuned for API latency (10ms to 5s range)
• Route labels: Group metrics by route pattern, not raw URL (avoids cardinality explosion)

---

See also

• Back to skills
• Architecture