Adding logging, metrics and error tracking to a service
Why it matters
Without monitoring you learn about incidents from your users. Structured logs, metrics and error tracking are how a team finds out first. Libraries are exempt — they have no runtime of their own.
How to fix it
Add the three signals:
• Errors: sentry-sdk
• Metrics: prometheus-client plus a /metrics endpoint
• Traces: opentelemetry-sdk
• Logs: structlog or loguru for structured JSON output
Does your repository pass this check?
Free for public GitHub repositories, no account needed. 50+ DevOps and security checks in about ten seconds.