Skip to main content

Documentation Index

Fetch the complete documentation index at: https://docs.foglamp.dev/llms.txt

Use this file to discover all available pages before exploring further.

A trace is one top-level generateText / streamText call. The Traces view lists them; opening one shows the full execution as a waterfall you can inspect span by span and replay in real time. See the data model for how traces and spans are defined.

Traces list

A paginated table (newest first) over the selected date range. Columns: trace id, name (traceName, falling back to agentName, else ), span count, tokens, duration, cost, and when it ran. Errored traces carry a red badge. Click a row to open the trace.

Waterfall

Each span is a row, indented by its parent (parentSpanId) so nested tool calls and steps sit under their parent. Span types are colour-coded:
TypeColourMeaning
agentamberthe root span — the whole call
llmvioletone model step
toolblueone tool execution
othergreyanything else
The bar’s offset and width are proportional to when the span started and how long it ran, so you can see serial vs. overlapping work at a glance.

Span detail

Click any span to open its detail panel:
  • Timing — start time, duration, and TTFT (time to first token).
  • Model — provider and model id.
  • Tokens & cost — input/output tokens, computed cost, and the pricing source that produced it (resolved price vs. a custom rule).
  • Throughput — for streaming LLM spans, a tokens/second headline.
  • Payloads — the captured input and output, pretty-printed and collapsible (subject to recordInputs/recordOutputs).
  • Errors & metadata — the error message (if any) and any span metadata.

Scores

If a trace has been scored by an eval, a Scores card lists each result as a badge — green for pass, red for fail, or the numeric score. Hover a badge to read the judge’s reason.

Replay

The Replay card plays the trace back on its real timeline: press play, scrub with the slider, and switch between 1× / 2× / 4× speed. As it plays, the readout shows the currently streaming span, its token count, and instantaneous tokens/second; the TTFT moment is marked on each LLM bar. Replay is powered by the intra-stream token samples the SDK records — no extra instrumentation needed. Opening a trace with ?replay=1 auto-plays it.