> For the complete documentation index, see [llms.txt](https://docs.rillet.com/EjmEP4KZ9BVp9j2ho4Mh/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.rillet.com/EjmEP4KZ9BVp9j2ho4Mh/aura-ai/aura-ai/audit-trail.md).

# Audit Trail

Audit Trail in **Rillet** shows you exactly how Aura generated a response or completed an action. You can review the data Aura used, the steps it followed, the logic applied, and the records or reports referenced to produce the final result. This is especially useful for accounting and finance workflows where teams need to validate calculations, review supporting data, or confirm how an AI-generated result was produced before approval.

### Understand How Audit Trail Works

After Aura completes a response, a **View reasoning** link appears below it. Audit Trail is available for requests where Aura performs calculations, retrieves financial data, generates analysis, or prepares actions or accounting entries. The diagram shows the path from a request to the supporting evidence.

<figure><img src="https://612265232-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FNpXFVFZpTQck3vIrjmHB%2Fuploads%2F2XcsM5kf6lyk9RlHlr1A%2F005-audit-trail.svg?alt=media&amp;token=c8976e2f-b8cb-4587-8a67-e2df0231c92c" alt=""><figcaption></figcaption></figure>

### Open the Audit Trail

Follow these steps to open the Audit Trail:

{% stepper %}
{% step %}
Open **Aura** from the side panel.
{% endstep %}

{% step %}
Send a request in an Assistant conversation.

<figure><img src="https://612265232-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FNpXFVFZpTQck3vIrjmHB%2Fuploads%2FFBS02LudfvjnG9utseQ4%2Fimage.png?alt=media&#x26;token=355dbe7f-ed2b-4a01-839c-f9400261506c" alt=""><figcaption></figcaption></figure>
{% endstep %}

{% step %}
Click the **View reasoning** link below the response.

<figure><img src="https://612265232-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FNpXFVFZpTQck3vIrjmHB%2Fuploads%2Fw5pI8oXrEtmmH6sCSHXy%2Fimage.png?alt=media&#x26;token=c1fcd1b9-db85-44b0-ace8-d01c12138efb" alt=""><figcaption></figcaption></figure>
{% endstep %}
{% endstepper %}

Aura opens the **Reasoning** page, a graph that shows how the response was generated, with an **Input** card for your request, one or more **Agent** cards, and an **Output** card. Click **Back to Assistant** to return to the conversation.

### Read the Reasoning Graph

The reasoning graph breaks the response into the individual steps Aura performed. Each node represents a sub-task Aura completed, such as retrieving data, validating information, performing calculations, generating summaries, or preparing records.

<figure><img src="https://612265232-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FNpXFVFZpTQck3vIrjmHB%2Fuploads%2FnvQhxNdfXG89tqCALA4q%2Fimage.png?alt=media&amp;token=46a831b1-51ca-4654-9dfb-042c13b06797" alt=""><figcaption></figcaption></figure>

| What the Graph Shows | Details                                                                                                                                                         |
| -------------------- | --------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| **Order**            | The order in which steps were executed                                                                                                                          |
| **Logic**            | The logic applied at each step                                                                                                                                  |
| **Links**            | Links to the supporting reports and records used during the process                                                                                             |
| **File**             | A file, such as a CSV, of the exact data state Aura had at the moment it executed, including the actual values it pulled and not just the query parameters used |

Use the file to verify the exact numbers behind an AI-generated entry, reconcile a discrepancy, or share supporting evidence with auditors. Many nodes include clickable links that open the underlying records directly from the reasoning view, making it easier to trace calculations back to their source data.

### Review an Example: Bad Debt Journal Entry

The following example shows how Audit Trail can help verify an AI-generated journal entry. Suppose you ask Aura to prepare a journal entry for bad debt expense. After Aura generates the entry, you can click **View reasoning** to review how it calculated the amount.

| Stage                            | What You May See                                                                                                                                                                                               |
| -------------------------------- | -------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| **Aura may**                     | Review overdue invoices, evaluate related credit memos, calculate the bad debt amount, create a proposed journal entry, and request approval before posting.                                                   |
| **The reasoning graph may show** | Validating the correct bad debt accounts, retrieving overdue invoices, reviewing total invoice balances, evaluating credit memo history, applying the calculation logic, and creating the journal entry draft. |

You can click into the related invoices or supporting records directly from the graph to verify the calculation details, so your team can confirm how the final amount was produced without manually recreating the analysis.

### Validate Results and Review Earlier Sessions

Use Audit Trail to review supporting calculations, validate financial analysis, trace values back to source records, and confirm the logic behind AI-generated outputs, especially before approving accounting entries or reviewing financial results. To look back at earlier sessions, open the **Audit trail** tab in **Activity**, which lists Assistant and Agent sessions with their **Category**, **Status**, **User**, and **Artifacts**. See [**Aura Activity**](/EjmEP4KZ9BVp9j2ho4Mh/aura-ai/aura-ai/aura-activity.md) for how to read it.

### See Also

To learn more about Audit Trail in **Rillet**, review these articles:

* [**Aura AI**](/EjmEP4KZ9BVp9j2ho4Mh/aura-ai/aura-ai.md)
* [**Getting Started with Aura Assistant**](/EjmEP4KZ9BVp9j2ho4Mh/aura-ai/aura-ai/getting-started-with-aura-assistant.md)
* [**Actions and Approvals**](/EjmEP4KZ9BVp9j2ho4Mh/aura-ai/aura-ai/aura-actions-and-approvals.md)
* [**Aura AI Prompt Examples**](/EjmEP4KZ9BVp9j2ho4Mh/aura-ai/aura-ai/aura-ai-prompt-examples.md)


---

# 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 by asking a question.

Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:

```
GET https://docs.rillet.com/EjmEP4KZ9BVp9j2ho4Mh/aura-ai/aura-ai/audit-trail.md?ask=<question>&goal=<user_goal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.

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.