Skip to main content
The Observability API exposes activity and measurements for investigation and monitoring. Different signals answer different questions about the same application.

​
Logs and metrics

Logs record individual messages and events. Project, time range, service, and trace filters help isolate relevant activity. Metrics describe measurements over time. Project metric queries use supported metric names and label filters; they do not accept arbitrary PromQL.

​
Traces and spans

A trace connects related operations across an execution. Spans describe parts of that work, including timing and error status. Trace summaries and details can return a bounded selection of spans. The span-list operation provides pagination and error filtering when you need to investigate further.

​
Runs and analytics

Execution API run events preserve execution progress and output. A trace adds diagnostic context; it does not replace the run record. Account analytics summarize conversation and run activity, including counts, outcomes, duration, and supported usage breakdowns. Spending and credit balances belong to the Billing and Usage API.

​
From a failure to a change

Suppose a support agent gives an incomplete answer after calling an external service. Its run events show the attempted work, a related trace can locate a failure or delay, and logs can supply the emitted error. Use shared request, trace, or run IDs where available. Nearby timestamps alone do not establish that records describe the same work, and permissions still apply to each resource. These findings can inform an implementation change and a new evaluation case. The Evaluations API provides the definitions and reports for assessing that change; creating a regression case is part of your development process.

​
Get started

Inspect logs and traces shows the queries and verification steps for investigating an execution.

​
API references