Skip to main content
The Trace View on the Dashboard breaks a session into a hierarchy of traces and spans for granular analysis. Use it to find where time goes in a turn and to debug errors across STT, LLM, TTS, and tool calls. The Trace View offers an even deeper level of insight, breaking down the entire session into a hierarchical structure of traces and spans.

​
Session Configuration

At the top level, the Session Configuration details every parameter the agent was initialized with: the STT, LLM, and TTS models, plus any function tools and MCP tools. Use it to reproduce and debug agent behavior.
Trace view hierarchy with the Session Configuration node selected, showing STT, LLM, and TTS models and function and MCP tools in the Properties panel

​
User & Agent Turns

The core of the Trace View is the breakdown of the conversation into User & Agent Turns. Each turn is a single exchange between the user and the agent, with a detailed timeline of these spans:
A user and agent turn expanded to show STT, EOU, LLM, and TTS processing spans along the timeline

​
Turn Properties

For each turn, inspect the properties of the components involved: the transcript of the user’s input, the response from the LLM, and any errors that occurred.
A turn selected in the trace view with its transcript and LLM response shown in the Properties panel

​
Tool Calls

When the LLM invokes a tool, the Trace View shows details about the call, including the tool’s name and the parameters it was called with. Use it to validate integrations and debug tool behavior.
A tool call span selected in the trace view, showing the tool name and the parameters passed to it

​
What’s Next

Observability Options

View per-turn traces and metrics on the Dashboard.

Session Analytics

See session-level metrics and transcripts.

​
References

​
Examples

Observability Hooks

Emit traces from a session.

​
SDK Reference

Metrics

Metrics in the Python API reference.