Monitor your agents
Astropods automatically monitors agents that interact with common AI model providers. After you deploy an agent, its token usage, tool calls, and latency appear on the Insights page. For deeper observability, Astropods provides adapters for common frameworks.
Agent health
An agent’s Monitor > Health page shows whether its application, memory, CPU, and activity are healthy, and when their alerts fired. Each category has a state (Normal, Warning, or Critical), its last 24 hours, and an alert history.
The alert history covers up to the last 30 days. Astropods rebuilds it from the cluster’s metrics with the same checks that raise alerts, so it includes alerts from before you first opened the page. On a cluster that keeps metrics for less than 30 days, the history covers only what the cluster still keeps.
The history shows when a condition was breached, not when a notification was sent. Astropods sends at most one notification a day for each alert, so a breach that sent no notification, such as a repeat later the same day, still appears in the history.
Framework guides
Manually attaching observability
When an adapter isn’t available for your framework, you can instrument an agent manually using standard OpenTelemetry. Every deployed agent container receives these environment variables automatically:
OTEL_EXPORTER_OTLP_ENDPOINT is unset during local development. Guard your instrumentation on the variable so the agent runs cleanly in both environments.
Node.js
Initialize the SDK
Initialize the SDK at the entry point of your agent, before any other imports that you want traced:
Emit spans
With the SDK running, emit spans wherever you want them — see the OpenTelemetry tracing API for the span API and the auto-instrumentation packages for HTTP, fetch, and other off-the-shelf coverage.