The Datadog APM Trace Analyzer skill connects to the Datadog Tracing API v2 to retrieve and analyze distributed trace data across microservice architectures. It queries trace search endpoints with faceted filters for service name, resource, duration percentiles, and error status to identify latency bottlenecks and high-error-rate endpoints. The skill generates flame graph summaries that highlight the slowest spans in each trace, calculates service-level p50/p95/p99 latency metrics, and maps upstream and downstream dependency impacts using the Datadog Service Map API. It configures custom trace metrics using the Datadog Metrics API for ongoing monitoring, sets up anomaly detection monitors for sudden latency spikes, and produces root cause analysis reports correlating trace degradation with recent deployments via the Datadog Events API. Output includes actionable recommendations with specific span-level optimization targets.