autonomy--sre-agent

To diagnose production issues, query logs, metrics, and traces from observability, cloud, and CI/CD CLIs to collect signals and form actionable root-cause hypotheses. The workflow checks error rates, correlates traces, and aligns deployments with incident timing.

The best way to investigate a service health check is applying a structured investigation workflow that queries observability data to assess service health, gather evidence, and test candidate root causes with safe remediation guidance.

To reproduce reported bugs with evidence, align deployment timing from CI/CD pipelines with cloud CLI logs and metrics to verify the incident context. This hypothesis-driven analysis generates ranked root causes supported by structured evidence.

Yes, you can use Datadog, AWS CLI, and GitHub CLI for root-cause analysis. The investigation workflow queries these observability, cloud, and CI/CD CLIs to correlate signals, validate hypotheses, and plan safe remediation.

You need hypothesis-driven analysis for incident investigations when complex production issues require querying multiple observability sources. It generates, ranks, and tests candidate root causes using collected logs and metrics to reduce mean time to resolution.