golang-samber-slog

The golang-samber-slog skill provides structured logging pipeline guidance and integration patterns for Go applications using the samber/slog-* ecosystem. It focuses on building composable logging architectures around Go 1.21+ `slog.Handler` interfaces, helping developers create scalable and maintainable log processing pipelines. The skill addresses common logging challenges such as high log throughput, noisy records, inconsistent formatting, backend routing, and sensitive data handling. It promotes a canonical pipeline model where sampling, formatting, routing, and sink delivery are ordered efficiently to reduce CPU usage and improve observability reliability.

The skill covers three primary libraries: `slog-multi` for handler composition and routing, `slog-sampling` for throughput and noise reduction, and `slog-formatter` for attribute transformation and sanitization. It also supports a wide ecosystem of integrations and middleware packages including HTTP frameworks like Fiber, Gin, Chi, and Echo, as well as backend sinks such as Datadog, Sentry, Loki, Syslog, Logstash, Graylog, Fluentd, Kafka, Slack, Telegram, and Microsoft Teams. Key capabilities include fanout pipelines, failover handlers, routing strategies, log sampling policies, PII filtering, error formatting, and backend-specific log delivery using standard `slog.Handler` implementations.

This skill is intended for Go developers, platform engineers, observability teams, and AI coding agents working on structured logging systems. It is especially useful in distributed services, microservices, APIs, and production systems where logs must be filtered, transformed, and routed to multiple destinations efficiently. Teams adopting Go's native `slog` package or already using any `github.com/samber/slog-*` library can use this skill to design cleaner logging architectures and integrate observability tooling consistently across services.