> For the complete documentation index, see [llms.txt](https://docs.wobby.ai/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.wobby.ai/ai-analysts/monitoring/traces.md).

# Traces

In Studio, the *Traces* tab gives [Admins](/settings/members.md) a full log of every conversation that has taken place with an AI Analyst. Each conversation — from a user asking a question to the analyst returning an answer — is recorded as a single trace. You can browse, filter, and open any trace to replay the full conversation.

{% hint style="info" %}
Only [Admins](/settings/members.md) can view all traces. [Users](/settings/members.md) can only see their own conversations.
{% endhint %}

## Accessing Traces

1. In Studio, go to *AI Analysts* and select the analyst you want to inspect.
2. Click the *Monitoring* button in the top action bar and select *Traces*.

The page has a two-panel layout:

* *Left panel* — a scrollable list of all traces for this AI Analyst, grouped by date.
* *Right panel* — the full conversation detail for whichever trace you have selected.

<figure><img src="/files/5tEazEQZ7tjt40vUWwr7" alt=""><figcaption><p>The Traces view showing a list of user sessions grouped by date</p></figcaption></figure>

## What Each Trace Shows

Each trace in the list displays:

| Field    | Description                                                |
| -------- | ---------------------------------------------------------- |
| User     | The name and avatar of the user who ran the session        |
| Time     | When the session took place (relative, e.g. "2 hours ago") |
| Question | The user's original question or prompt                     |
| Channel  | Where the question was asked — Web, Slack, or Teams        |
| Feedback | Whether the user rated the answer thumbs up or thumbs down |
| Comments | Number of comments or annotations on the trace             |

## Filtering Traces

Use the filter bar at the top of the left panel to narrow down the list. Available filters:

* *Feedback* — filter by positive, negative, or no feedback
* *Follow-ups* — show only sessions that included follow-up questions
* *Assumptions* — show only sessions where the analyst made assumptions
* *Channel* — filter by Web, Slack, or Teams
* *Starred* — show only traces you have starred
* *User* — filter by a specific user
* *Duration* — filter by how long the session took

## Viewing a Trace

Click any trace in the left panel to open the full conversation in the right panel. You will see:

* The user's original question and all follow-up messages
* Every response from the AI Analyst, including reasoning steps, charts, and tables

This is useful for understanding how the analyst reasoned through a question, identifying where it may have gone wrong, and reviewing the quality of answers before acting on any feedback.

## Starring Traces

Click the star icon on a trace to mark it for easy reference later. You can then filter the list to show only starred traces.

## Exporting Traces

Click *Export CSV* to download the current filtered list of traces as a spreadsheet. The export includes the question, user, timestamp, channel, duration, and feedback for each trace.

{% hint style="info" %}
The export button has a short cooldown after each download to prevent duplicate exports.
{% endhint %}


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter:

```
GET https://docs.wobby.ai/ai-analysts/monitoring/traces.md?ask=<question>&goal=<endgoal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.