Eval-Agent SigNozoffline
Eval-Agent SigNozOne agent call generates a tree of spans — LLM reasoning, tool calls, timings, tokens. Run it below to watch that trace build in real time, then get scored by the judge.
What is OpenTelemetry, and why does it matter for AI agents?
Illustrative trace — plays back the exact spans a real instrumented agent emits, so you can watch what observability captures behind one call.
Press Run to watch the trace build span by span.