speckit.sre-observabilitylisted

## 🎯 Mission Design, configure, and enforce enterprise-grade Observability and Site Reliability Engineering (SRE) standards across services. Ensure system state is fully transparent through the Three Pillars (Logs, Metrics, Traces), actionable alerts, and battle-tested resilience patterns. ## 📥 Required Inputs - `.agents/specs/[feature]/spec.md`, `plan.md`, and `tasks.md` - Service architecture, topology, and SLA/SLO commitments - Infrastructure configs (`docker-compose.yml`, Kubernetes manifests, Helm charts) ## 📋 Protocol ### 1. Distributed Tracing & OpenTelemetry (OTel) - **Trace Context Propagation**: Propagate W3C TraceContext headers (`traceparent`, `tracestate`) across all HTTP, gRPC, and message queue boundaries. - **Span Granularity**: Instrument high-level operations: incoming HTTP request, external API calls, DB transactions, background jobs, and cache accesses. - **Baggage & Correlation**: Attach `trace_id`, `span_id`, `service.name`, and `environment` to all telemetry data. ### 2. Metrics Architecture (RED & USE Methods) - **RED Method (Request-driven Services)**: - *Rate*: Requests per second (Counter). - *Errors*: Failed requests count partitioned by HTTP status code / error type (Counter). - *Duration*: Request latency distribution (Histogram with standardized buckets: `0.005s, 0.01s, 0.025s, 0.05s, 0.1s, 0.25s, 0.5s, 1s, 2.5s, 5s, 10s`). - **USE Method (Resources - CPU, RAM, Disk, DB Pools)**: - *Utilization*: Percent time busy. - *Saturati