← Back to Discover

DigitalOcean Insights

DigitalOcean Insights is built-in observability for agents, applications, and infrastructure running on DigitalOcean, currently in public preview.

Observability & evalsApplication & server monitoring

What it does

DigitalOcean Insights brings metrics, logs, and traces from supported DigitalOcean products into one backend. It is part of DigitalOcean Agent Learning Services. Telemetry from DigitalOcean Kubernetes (DOKS), App Platform, Managed Agents, and other products flows in through built-in integrations. Users do not need to install, deploy, or instrument anything. Insights runs on OpenTelemetry, an open standard for observability data. The telemetry shares dimensions such as resource, project, and tag. Users open Insights in the control panel. They browse signals in Explorers, build charts with a visual builder or PromQL, and save charts to dashboards. They can also set alerts on metrics.

Capabilities

  • Metrics and logs for Serverless Inference, Batch Inference, Inference Router, Managed Agents, and Action Gateway show token throughput, latency, error rates, and cost estimates.
  • Action Gateway adds traces that follow a single agent turn from the model call through each tool it invoked.
  • Alerts use a condition on any metric, an evaluation window, and a re-notify interval while the alert stays unresolved.
  • The public API manages alerting and notification channels and queries metrics, logs, and traces programmatically.
  • Role-based access control manages who on a team can view and configure Insights.

When to use it

  • An agent run slows down or fails, and the cause could be anywhere in the chain of model calls, routing decisions, tools, and infrastructure.
  • A metric spikes on an application running on DigitalOcean Kubernetes or App Platform, and the related logs are in another tool.
  • A team wants to connect metrics, logs, and traces without building integrations or instrumenting its stack with a third-party observability tool.