> ## Documentation Index
> Fetch the complete documentation index at: https://docs.egma.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# What is monitoring?

**Monitoring** collects evidence from your agent's production conversations.
Use it to inspect what callers said, how the agent replied, which tools it
called, and the available performance metrics.

You do not need tests or simulation runs to monitor an agent. LiveKit sends
production traces through the Egma SDK. Retell imports completed voice calls
through its API.

## Inspect a conversation

[Review production conversations](/docs/platform/monitoring/review-conversations) to read transcripts, metrics, and grades in **Traces**.

## Set up monitoring

Follow [Set up monitoring](/docs/platform/monitoring/set-up-monitoring) for Retell
or LiveKit. Complete a real conversation and verify that its evidence arrives
in your project before treating the setup as complete.

Read [Monitoring philosophy](/docs/core-philosophies/monitoring-philosophy) for how to use
production findings to improve your tests.