Observability

Catalyst emits metrics, distributed traces, and structured logs for every workflow, agent, and Dapr API call — without requiring instrumentation in application code.

Metrics

Request rates, error rates, and latency percentiles per App ID, surfaced in the Metrics console page and exportable to external backends (Prometheus, Datadog, Grafana).

Distributed traces

Every Dapr API call carries trace context. Traces flow across App IDs and into backing infrastructure, giving end-to-end visibility of multi-service requests. Configure trace sampling and exporter targets in Observability settings.

API logs

A structured log of every Dapr API call — method, status, latency, and App ID — available in the API Logs console page. For AI agents, LLM calls (via the Conversation API) are included with token counts.

App Graphs

The App Graph visualizes the live call topology across your project — who called whom, how many times, and at what error rate. Use it to trace requests and identify hotspots.

For configuration (trace backends, sampling, log retention), see Observability.