What problem does it solve?
This skill helps teams quickly understand and triage issues by querying and analyzing Datadog logs, metrics, and APM data. It provides scripted workflows and a reference for Datadog query language.
Core Features & Use Cases
- Statistics-first diagnostics: run structured statistics to gauge event volume, error rate, top services, and recurring patterns before collecting raw logs.
- Strategic log sampling: fetch targeted logs to accelerate root-cause analysis without overwhelming data volumes.
- Query guidance: built-in Datadog Query Language (DQL) references and templates for common filters (service, host, status, time ranges) and aggregation.
Quick Start
Run the statistics-first workflow to establish context:
- python .claude/skills/observability-datadog/scripts/get_statistics.py --time-range 60
Then fetch targeted logs:
- python .claude/skills/observability-datadog/scripts/sample_logs.py --strategy errors_only --service my-service
Optional: explore with the Datadog client:
- python -c "from datadog_client import search_logs; print(search_logs('service:my-service', time_range_minutes=60, limit=20))"