> ## Documentation Index
> Fetch the complete documentation index at: https://docs.roserx.ai/llms.txt
> Use this file to discover all available pages before exploring further.
# Audit Log
> Track every action taken across the RoseRx Platform with a comprehensive, searchable audit trail.
The Audit Log provides a comprehensive trail of every significant action performed across the RoseRx Platform. It supports your governance and compliance requirements by recording who did what, when, and whether the action succeeded — giving you full accountability across your organization.
The Audit Log captures actions automatically. No configuration is required — every tracked action is logged from the moment your organization is created.
## Accessing the audit log
Navigate to **Monitor → Audit Log** in the left sidebar.
## Filtering events
Use the filter bar to narrow the log to the events you're interested in:
* **Actor** — Filter by the person or system that performed the action.
* **Action** — Filter by the type of action (e.g. created, updated, deleted).
* **Target Type** — Filter by what the action was performed on (e.g. Agent, Widget, Document).
* **Result** — Filter by outcome: Success, Failure, or Denied.
* **Date range** — Select a custom date range to focus on a specific period.
## Audit log table
Each row in the table represents a single action, displayed with the following columns:
| Column | Description |
| ------------- | -------------------------------------------------------------------------- |
| **Timestamp** | The exact date and time the action occurred |
| **Actor** | Who performed the action, with a color-coded type badge |
| **Action** | What was done (e.g. "created", "updated", "deleted") |
| **Target** | The specific item the action was performed on |
| **Result** | Color-coded outcome: green for Success, red for Failure, yellow for Denied |
### Actor types
Actions can be performed by different types of actors, each identified by a color-coded badge:
| Actor Type | Badge Color | Description |
| ---------- | ----------- | -------------------------------------------------- |
| **User** | Blue | A human team member performed the action |
| **System** | Purple | An automated platform process performed the action |
| **API** | Orange | The action was triggered via the API |
## Expandable row details
Click any row to expand it and see additional information:
* **Description** — A human-readable summary of what happened.
* **Metadata** — Key-value pairs providing additional context, such as changed field names, previous values, or related entity IDs.
## Actions tracked
The Audit Log records actions across every major area of the platform, organized by category:
* **created** — A new agent was created.
* **updated** — Agent settings were modified (e.g. model, tone, response length).
* **published** — An agent was published and made active.
* **archived** — An agent was archived and removed from active use.
* **deleted** — An agent was permanently deleted.
* **guidance\_updated** — Guardrails or guidance settings were changed.
* **created** — A new widget was created.
* **updated** — Widget settings were modified (e.g. style, tabs, domain restrictions).
* **deleted** — A widget was deleted.
* **created** — A new deployment was created.
* **updated** — Deployment settings were modified.
* **published** — A deployment was published and made active.
* **deleted** — A deployment was deleted.
* **uploaded** — A document was added to an agent's knowledge base.
* **updated** — A document's details were modified.
* **deleted** — A document was removed from the knowledge base.
* **reindexed** — A document was reprocessed and re-indexed.
* **URL added** — A URL source was added to the knowledge base.
* **uploaded** — A resource file was uploaded.
* **created** — A resource link was created.
* **deleted** — A resource was removed.
* **config updated** — Adverse event monitoring configuration was changed.
* **status updated** — An adverse event's review status was updated (individually or in bulk).
* **reviewed** — An adverse event was reviewed and notes were added.
* **closed** — An adverse event was closed.
* **exported** — Adverse event data was exported (CSV or regulatory format).
* **created** — A new topic was created (individually or in bulk from templates).
* **updated** — A topic's keywords or settings were modified.
* **deleted** — A topic was removed.
* **created** — A new batch test was created.
* **deleted** — A batch test was deleted.
* **duplicated** — A batch test was duplicated.
* **questions added / deleted** — Questions were added to or removed from a test.
* **submitted** — A widget visitor submitted feedback on an agent response.
* **invited** — A new team member was invited to the organization.
* **removed** — A team member was removed from the organization.
* **role changed** — A team member's role was updated.
* **updated** — Organization or platform settings were modified.
## Exporting the audit log
Click the **Export CSV** button to download audit log data. Exports can include up to **10,000 events** based on your current filters.
Apply filters before exporting to keep the CSV focused and manageable. For compliance audits, use the date range filter to pull events for the specific reporting period.
## Common use cases
Track who made changes to agent guardrails, adverse event settings, or other compliance-critical configuration. Export the log for external audit purposes.
If an agent's behavior changes unexpectedly, use the audit log to trace recent modifications and identify the cause.
Review user actions to ensure appropriate access and usage patterns across your team.
Monitor automated system actions such as document reindexing or scheduled processes.