> For the complete documentation index, see [llms.txt](https://ai-security-docs.akto.io/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://ai-security-docs.akto.io/akto-atlas-agentic-ai-security-for-employee-endpoints/ai-agent-activity/audit-data-akto-atlas.md).

# Audit Data - Akto Atlas

## Overview

The **Audit Data** page in **Akto Atlas** shows you every MCP server, skill, LLM, AI agent, and vendor your employees use, and lets you **control how they're allowed to operate**.

From here, you can:

* See which MCP servers, skills, LLMs, AI agents, and vendors are being accessed.
* Inspect the tools, resources, and prompts exposed by each MCP server.
* Approve, block, or conditionally allow any of them, individually or in bulk.
* Scope any block to specific teams, devices, or users instead of blocking everyone.

## Explore Audit Data

Use the tabs at the top of the page to switch between **MCP Servers**, **Skills**, **LLMs**, **AI Agents**, and **Vendors**. Each tab shows a live count, and you can filter the time range with the **All time** dropdown at the top right.

{% tabs %}
{% tab title="MCP Servers" %}
Each row represents an MCP server and the agent that accesses it.

<div data-with-frame="true"><figure><img src="https://1895868029-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FkpBG2Nfd5ZbNmahr8W6H%2Fuploads%2Fgit-blob-63cecd688a38cdc5f77518dd5461503894fc25dc%2Faudit-data-mcp-servers.png?alt=media" alt="MCP Servers tab in Audit Data with two servers selected" width="563"><figcaption><p>Select MCP servers to allow, block, or add to your MCP registry.</p></figcaption></figure></div>

The table has the following columns:

<table><thead><tr><th width="147.92578125">Column</th><th>What it tells you</th></tr></thead><tbody><tr><td><strong>MCP Server</strong></td><td>The MCP server (domain or endpoint) being accessed</td></tr><tr><td><strong>AI Agent</strong></td><td>The agent accessing the server (e.g. VSCode, Claude, Cursor)</td></tr><tr><td><strong>Last Detected</strong></td><td>When the server was first observed in Atlas</td></tr><tr><td><strong>Updated</strong></td><td>Most recent activity for this server</td></tr><tr><td><strong>Access Type</strong></td><td>Type of access (e.g. public, private, third-party)</td></tr><tr><td><strong>Remarks</strong></td><td>Current decision: <strong>Approved</strong>, <strong>Rejected</strong>, or <strong>Conditionally Allowed</strong>. A clock icon next to the remark means an <strong>Audit Pending</strong> re-review is due.</td></tr><tr><td><strong>Marked By</strong></td><td>Who last updated the decision</td></tr></tbody></table>

{% hint style="info" %}
**Default Approval Behaviour**

* Without an [MCP registry integration](/integrations/mcp-registry-integration.md) set up, every newly discovered MCP server defaults to **Approved**.
* Once you set up an [MCP registry integration](/integrations/mcp-registry-integration.md), every newly discovered MCP server instead defaults to **Blocked**. From there, you either conditionally allow it for a specific agent (Cursor, Claude, and so on), or [add it to your MCP registry](#action-dropdown), which gets it approved by default.
  {% endhint %}

You can act on servers directly from the table, without opening the full server details:

* **Bulk actions:** Select one or more servers using the checkboxes. The action bar at the bottom of the table offers **Allow \<N> selected servers**, **Block \<N> selected servers**, and **Add to MCP registry**.
* **Quick actions:** Hover a row to open its quick actions (**Allow this server**, **Block this server**, **Add to MCP registry**).
  {% endtab %}

{% tab title="Skills" %}
Each row represents a skill discovered on your employees' agentic tools.

<div data-with-frame="true"><figure><img src="https://1895868029-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FkpBG2Nfd5ZbNmahr8W6H%2Fuploads%2Fgit-blob-b4626729c9310c24834601eac742b4c59b52e1d0%2Fimage%20(230).png?alt=media" alt="" width="563"><figcaption></figcaption></figure></div>

<table><thead><tr><th width="147.92578125">Column</th><th>What it tells you</th></tr></thead><tbody><tr><td><strong>Skill</strong></td><td>The name of the skill</td></tr><tr><td><strong>Source</strong></td><td>The agent the skill was discovered on (e.g. codex, claude), or <strong>not-attached</strong> if it isn't tied to a specific agent</td></tr><tr><td><strong>Last Detected</strong></td><td>When the skill was first observed in Atlas</td></tr><tr><td><strong>Remarks</strong></td><td>Current decision: <strong>Allowed</strong>, <strong>Blocked</strong>, or <strong>Conditionally Allowed</strong></td></tr><tr><td><strong>Marked By</strong></td><td>Who last updated the decision</td></tr></tbody></table>

{% hint style="success" %}
**Bulk Blocking**

Select one or more skills using the checkboxes, then click **Block \<N> selected skills** from the action bar at the bottom of the table to block them all in one action.
{% endhint %}
{% endtab %}

{% tab title="LLMs" %}
The **LLMs** tab lists the LLM services your employees reach, shown by domain (for example chatgpt.com, claude.ai, perplexity.ai).

<div data-with-frame="true"><figure><img src="https://1895868029-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FkpBG2Nfd5ZbNmahr8W6H%2Fuploads%2Fgit-blob-263321fdf9589c4665c1b7cbe07a546eb68095b2%2Faudit-data-llms.png?alt=media" alt="LLMs tab in Audit Data with four LLMs selected" width="563"><figcaption><p>Select LLMs and block them from the action bar.</p></figcaption></figure></div>

Each row shows when the LLM was last detected, its most recent activity, its current decision (for example **Allowed**), and who last updated the decision.

To block LLMs, select one or more rows using the checkboxes, then click **Block \<N> selected LLMs** from the action bar at the bottom of the table. Guardrails then stop prompts sent to the selected LLMs for the people you choose, as described in [Scope a Block to Specific Teams, Devices, or Users](#scope-a-block-to-specific-teams-devices-or-users).
{% endtab %}

{% tab title="AI Agents" %}
The **AI Agents** tab lists the AI agents (for example Claude CLI, Claude Desktop, Cursor, Codex, VS Code, Windsurf) and local models (for example llama3.2:3b, qwen3:0.6b) detected across your employees' devices.

<div data-with-frame="true"><figure><img src="https://1895868029-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FkpBG2Nfd5ZbNmahr8W6H%2Fuploads%2Fgit-blob-0b2a0a82105f850558fb1908ebfc9a2150375359%2Faudit-data-ai-agents.png?alt=media" alt="AI Agents tab in Audit Data with two agents selected" width="563"><figcaption><p>Select AI agents and block them from the action bar.</p></figcaption></figure></div>

Each row shows when the agent was last detected, its most recent activity, its current decision (for example **Allowed**), and who last updated the decision.

To block agents, select one or more rows using the checkboxes, then click **Block \<N> selected AI agents** from the action bar at the bottom of the table. Guardrails then stop prompts and MCP calls from the selected agents for the people you choose, as described in [Scope a Block to Specific Teams, Devices, or Users](#scope-a-block-to-specific-teams-devices-or-users).
{% endtab %}

{% tab title="Vendors" %}
The **Vendors** tab lists the AI vendors (for example Anthropic, OpenAI, Google, AWS, GitHub Copilot) detected across your employees' devices.

<div data-with-frame="true"><figure><img src="https://1895868029-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FkpBG2Nfd5ZbNmahr8W6H%2Fuploads%2Fgit-blob-31b39e95d2a3b86fe50a4fd0852c689f08b5f331%2Faudit-data-vendors.png?alt=media" alt="Vendors tab in Audit Data with two vendors selected" width="563"><figcaption><p>Select vendors and add them to the allowed list.</p></figcaption></figure></div>

Each row shows the vendor, a count of its activity, and its current status. Every vendor starts as **Not approved**.

To approve vendors, select one or more rows using the checkboxes, then click **Add \<N> selected vendors to allowed list** from the action bar at the bottom of the table.
{% endtab %}
{% endtabs %}

### Scope a Block to Specific Teams, Devices, or Users

On every tab (**MCP Servers**, **Skills**, **LLMs**, **AI Agents**, and **Vendors**), clicking **Block** opens a dialog that asks **Who is blocked**. The scope you choose limits the block to specific people instead of everyone:

<div data-with-frame="true"><figure><img src="https://1895868029-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FkpBG2Nfd5ZbNmahr8W6H%2Fuploads%2Fgit-blob-87ced61b0d1ee51a797f8e45943b747d1b9d42cc%2Faudit-data-block-ai-agents.png?alt=media" alt="Block 2 AI agents dialog with Select Device Tags &#x26; Users chosen and a Groups condition" width="563"><figcaption><p>Choose who the block applies to, then add conditions to narrow the targets.</p></figcaption></figure></div>

* **Apply to all** (recommended): Blocks the selected items for every user in your organization. The dialog shows the user count.
* **Select Device Tags & Users**: Blocks the selected items only for the people you target. You build the rule as **Where my \<attribute> are \<value>**, and **Add condition** combines more than one rule. **Clear all** removes every condition.

You can target by:

| Attribute | Blocks it for                 |
| --------- | ----------------------------- |
| Groups    | Users in the selected groups  |
| Role      | Users with the selected roles |
| Teams     | Users in the selected teams   |
| Users     | Individual users (**Beta**)   |

As you build the rule, the dialog shows how many devices the selection covers (for example, **Applies to 12 Devices**) and which groups it includes. Click **Block selected users** to enforce the rule.

{% hint style="warning" %}
**User targeting is in beta.** Targeting individual users is currently supported for browser extensions and Claude Desktop.
{% endhint %}

## View MCP Server Details

Click on any MCP server to view its details, including all tools, resources, and prompts it exposes.

<div data-with-frame="true"><figure><img src="https://1895868029-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FkpBG2Nfd5ZbNmahr8W6H%2Fuploads%2Fgit-blob-69cb4dcd15811587632b636b5963115a04d0dd74%2Fimage%20(85).png?alt=media" alt="" width="563"><figcaption></figcaption></figure></div>

### Capabilities

Each MCP server exposes capabilities used by agents. For each capability, you can see:

<table><thead><tr><th width="224.94921875">Field</th><th>What it tells you</th></tr></thead><tbody><tr><td><strong>Type</strong></td><td>Whether it's a Tool, Resource, or Prompt</td></tr><tr><td><strong>Risk Analysis</strong></td><td>Any risk signals like <strong>Privileged Access</strong> or <strong>Malicious</strong></td></tr><tr><td><strong>Name</strong></td><td>The identifier of the capability</td></tr><tr><td><strong>Access Types</strong></td><td>Whether the capability is public, private, or third-party</td></tr><tr><td><strong>Remarks</strong></td><td>Whether it's <strong>Approved</strong>, <strong>Rejected</strong>, or <strong>Conditionally Allowed</strong></td></tr><tr><td><strong>Marked By</strong></td><td>Who made the decision</td></tr></tbody></table>

### Access Control Options

You can set access decisions at both the server level (via the **Action** dropdown) and the individual tool level. Use the following options:

#### **Allow**

Grant full access to the MCP server or specific tool. The agent can use all capabilities without restrictions.

#### **Block**

Deny access to the MCP server or specific tool. After you click **Block**, the dialog asks **Who is blocked**, and you choose whether the block applies to everyone or only to specific teams, devices, or users, as described in [Scope a Block to Specific Teams, Devices, or Users](#scope-a-block-to-specific-teams-devices-or-users).

### Action Dropdown

From the **Action** dropdown at the top of the server details view, you can make server-level decisions:

* [**Allow this server**](#allow) – Grant full access to all capabilities
* [**Block this server**](#block) – Deny all access immediately for the particular AI Agent
* **Block for all agents** – Block this server across all agents in your organisation
* [**Conditionally allow this server**](#setting-conditional-approval) – Grant access with defined restrictions
* **Add to MCP registry** – Register the server in your organisation's MCP registry

  <div data-with-frame="true"><figure><img src="https://1895868029-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FkpBG2Nfd5ZbNmahr8W6H%2Fuploads%2Fgit-blob-7df42e544de6deea17aa0fe897f2a2ae56241994%2Fimage%20(100).png?alt=media" alt="" width="563"><figcaption></figcaption></figure></div>

{% hint style="info" %}
To use **Add to MCP registry**, you must first set up an [MCP registry integration](/integrations/mcp-registry-integration.md) from **Settings → Integrations → MCP Registry**.
{% endhint %}

#### Setting Conditional Approval

If you choose **Conditionally Allow**, you can set clear boundaries for how the component is allowed to operate. The following components can be set:

{% tabs %}
{% tab title="Time Duration Allowed" %}
You define how long the component can remain active. Once the duration expires, Akto automatically blocks it.

<figure><img src="https://1895868029-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FkpBG2Nfd5ZbNmahr8W6H%2Fuploads%2Fgit-blob-35cffc608f793139ccfe433ff202cb98d3ff5c45%2FScreenshot%202025-12-17%20at%207.27.30%E2%80%AFPM.png?alt=media" alt="" width="563"><figcaption></figcaption></figure>
{% endtab %}

{% tab title="IPs Allowed" %}
You control where the component can be used from. You can allow:

* All IPs
* Specific IPs
* An IP range (CIDR)

<figure><img src="https://1895868029-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FkpBG2Nfd5ZbNmahr8W6H%2Fuploads%2Fgit-blob-317d4661424e47b9c5957f592f5667c04aea1077%2FScreenshot%202025-12-17%20at%207.27.52%E2%80%AFPM.png?alt=media" alt="" width="563"><figcaption></figcaption></figure>
{% endtab %}

{% tab title="Endpoints Allowed" %}
You choose which endpoints the component can access.

<figure><img src="https://1895868029-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FkpBG2Nfd5ZbNmahr8W6H%2Fuploads%2Fgit-blob-b8df1e973695ab1c1951cd6de42546a85e057eb6%2FScreenshot%202025-12-17%20at%207.28.02%E2%80%AFPM.png?alt=media" alt="" width="563"><figcaption></figcaption></figure>
{% endtab %}

{% tab title="Justification" %}
You add a mandatory justification so your team understands why you approved the component with conditions.

<figure><img src="https://1895868029-files.gitbook.io/~/files/v0/b/gitbook-x-prod.appspot.com/o/spaces%2FkpBG2Nfd5ZbNmahr8W6H%2Fuploads%2Fgit-blob-1cafe1c2a0ad00af55ebcf7ea1a66a8fe7a8a708%2Fimage%20(3)%20(1)%20(1)%20(1)%20(1)%20(1)%20(1)%20(1)%20(1)%20(1)%20(1)%20(1).png?alt=media" alt="" width="563"><figcaption></figcaption></figure>
{% endtab %}
{% endtabs %}

After configuring everything, click **Approve with Conditions** to enforce the restricted access.