# Endpoint Agents

{% hint style="warning" %}
Due to recent platform-wide naming, navigation, and URL changes in the ThousandEyes product, you may notice some discrepancies between the product and the screenshots displayed in our technical documentation. The instructions and actual pages in the product are still valid and haven’t changed. Please bear with us as we update our screenshots to better match the in-product experience. See the full scope of changes on [Naming and Navigation Menu changes - Summary List](https://docs.thousandeyes.com/whats-new/naming-and-nav-phase-2-changes).
{% endhint %}

The ThousandEyes Endpoint Agent is an application that is installed on Windows or macOS machines to collect network- and application-layer performance data when users access specific websites from within monitored networks.

## Introduction to Endpoint Agents

Endpoint Agents allow customers to leverage regular end-user desktops and laptops as test devices for measuring internal and external application performance. As Endpoint Agents can be deployed on and off-premise to readily available machines in any location or network, performance measurements can be run from almost every possible user environment.

Deploying Endpoints gives you the ability to identify and diagnose end-user issues in near real-time. Unlike [Enterprise Agents](https://docs.thousandeyes.com/product-documentation/thousandeyes-basics/working-with-agent-settings), Endpoint Agents run natively on Windows and Mac OSX operating systems and can be deployed on mass using standard automated deployment tools. Figure 1 below shows a typical example of deployment locations for Endpoint, [Enterprise](https://docs.thousandeyes.com/product-documentation/thousandeyes-basics/working-with-agent-settings) and [Cloud](https://www.thousandeyes.com/product/cloud-agents) Agents.

**Figure 1. Example Endpoint, Enterprise and Cloud Agent deployment locations**

![](/files/-M5xtURhPFZXvx97TCLh)

Endpoint Agents include the following features that are responsible for measuring the application performance:

* **Synthetic Tests**: A synthetic test template combines scheduled and dynamic tests to monitor a specific application. All the major applications have a defined synthetic test template, and you can also create a custom template for applications that do not have a predefined template.
  * **Scheduled Tests** - Scheduled tests run from Endpoint Agents at regularly scheduled intervals. These tests run without user interaction and can be deployed on Network (agent-to-server tests) and Web (HTTP server tests) layers.
  * **Dynamic Tests** - Dynamic Tests are similar to Scheduled Tests but, in addition they allow you to configure an Endpoint Agent to dynamically create tests to IP addresses and ports it detects when an application opens a connection.
* **Real User Tests** - Endpoint Agents record in-browser and network performance metrics whenever users navigate to monitored domains from monitored networks.
* **Network Access** - This feature records the performance of the Endpoint's physical wired or wireless connections, gateways, VPNs, proxies, and DNS servers.

You have full control of when and under what conditions the tests run. For example, a Scheduled Test deployed to a laptop can be switched on or off as it (the laptop) transits through different locations, wireless and wired networks, VPNs, and proxies. Similarly, the Real User Test recordings will only occur when the Endpoint Agent is on a specified network and also limited by configured target domains.

All the collected data and metrics are displayed in one intuitive **Endpoint Experience > Views** page. The Views page separates Scheduled Tests, Dynamic Tests, Real User Tests, and Network Access data by layers and the interactive timeline presents performance data that can be used to rapidly search through historical data. An example of the Views page, the interactive timeline, layers, metrics, and filters can be seen in Figure 2 below. With your Endpoint Agent data, you can create proactive alerts, email, and create highly customizable dashboards. All Endpoint Agent data is available via our [API](https://developer.cisco.com/docs/thousandeyes/v7/endpoint-agents-api-overview/).

**Figure 2. An example of just one of the many ways interactive data is presented:** **Endpoint Experience > Views**

![](/files/-M5xtURkAhZIvbJGx6jh)

There are two types of Endpoint Agents: **Endpoint Agent** and **Endpoint Agent Pulse**. The key difference between the two is that Endpoint Agent Pulse does not include the real user test feature. Throughout this article and the subsequent articles, all topics discussed apply to both products unless explicitly mentioned.

This article further explains: where and how to deploy Endpoint Agents, configuration options, how to run tests, end-user tools and experience, how to navigate your way through the UI and the layers of test views, how to create dashboards, how to troubleshoot deployments and operations, user permissions and much more.

## Deploying Endpoint Agents

Typically, organizations have multiple teams that manage their IT estate so to assist in the deployment and planning stages and to ensure the right teams are involved do check the [pre-deployment requirements](https://docs.thousandeyes.com/product-documentation/global-vantage-points/endpoint-agents/installing/system-requirements). the teams who manage functions such as software deployments, end-user access control, VPN-access, proxies and anti-virus systems must be involved as early as possible, so the Endpoint deployment runs smoothly. Use the following links for detailed instructions on how to install Endpoint Agents.

* [Downloading the Endpoint Agent Installer](https://docs.thousandeyes.com/product-documentation/global-vantage-points/endpoint-agents/installing/download-the-installer)
* [Installing the Endpoint Agent](https://docs.thousandeyes.com/product-documentation/global-vantage-points/endpoint-agents/installing/install-the-endpoint-agent)
* [Installing the Browser Extension](https://docs.thousandeyes.com/product-documentation/global-vantage-points/endpoint-agents/installing/install-the-browser-extension)
* [Installing Endpoint Agent for Windows OS via Group Policy](https://docs.thousandeyes.com/product-documentation/global-vantage-points/endpoint-agents/installing/install-endpoint-agent-for-windows-via-group-policy)

## Endpoint HTTP proxy support

The Endpoint Agent (and pulse) support the use of HTTP proxies. The network performance from an Endpoint Agent to a proxy is automatically measured during Scheduled Tests (as a part of synthetic tests) or Real User Test results. The network performance tests consist of a hop-by-hop path trace and end-to-end latency, loss, and jitter tests. Network tests are unable to reach the final target when a proxy is in use (Figure 3 below shows a proxy in use and the Network layer tests unable to reach the final target), therefore, all network tests are performed to the proxy instead.

### Real User Tests proxy

The proxy used by Browser Sessions is not configurable but automatically detected from the host system. See Figures 3 and 4 below for examples of how we present proxy data.

**Figure 3. Details view showing proxy measurements:** **Endpoint Experience > Views > Real User Tests > Visited Pages > Sessions**

![](/files/-M5xtURqgCDdps6wLYf9)

**Figure 4.** **Endpoint proxy displayed in Path Visualisation view:** **Endpoint Experience > Views > Real User Tests - . Network > Path Visualization**

![](/files/-M5xtURwKJRCOoszQHIe)

### Scheduled Tests proxy

You can configure and store multiple proxy configurations and apply them to all tests or set a unique per-test proxy. The proxy configuration supports static assignments, PAC files, and Basic and NTLM proxy authentication methods. To apply a saved proxy configuration for all scheduled tests use the settings on the **Endpoint Experience > Agent Settings** page or for the per-test proxy scenario apply a proxy configuration on the **Endpoint Experience > Test Settings** page. The Endpoint Agent supports only proxies that are used to handle HTTP and HTTPS traffic. The connection from the Endpoint Agent to the proxy is established using HTTP. You can learn more about this [here](https://www.thousandeyes.com/blog/measuring-performance-with-http-proxies)

For detailed instructions on proxy configurations and options see the [Proxy Configuration](https://docs.thousandeyes.com/product-documentation/endpoint-agent/endpoint-agent-proxy-configuration-for-scheduled-tests) article. For more details on how to navigate through the Views page to see your proxy measurements head to the Test, Real User Test, and Network Access Views and metrics section.

## Endpoint Agent updates

We usually ship Endpoint Agent updates every 2 weeks for both the Windows and Mac OSX operating systems. The Endpoint applications have auto-update process that runs at regular intervals to check, download, and install package updates. The Chrome browser extension automatically updates whenever a new Endpoint version is published in the Chrome Store. For full details on how the Endpoint Agent update process works see the [How does the Endpoint Agent work](https://docs.thousandeyes.com/product-documentation/endpoint-agent/how-does-the-endpoint-agent-work) article.

## Endpoint Agent Permissions and PII Management

ThousandEyes provides granular control over endpoint data through a set of dedicated permissions. These permissions allow administrators to manage access to sensitive information while ensuring teams can continue to troubleshoot effectively. You can configure these permissions by creating custom roles; for more information, see the [Role-based Access Control article](https://docs.thousandeyes.com/product-documentation/user-management/authorization/rb-access-control/role-based-access-control-explained).

### Available Endpoint Permissions

The following permissions control access to identifying information:

* **View endpoint data that identifies Endpoint Agents:** Controls access to identifying agent information, including the agent name, hostname, and serial number. Users without this permission will see a numeric ID instead of the agent name.
* **View endpoint data that identifies users:** Controls access to the username on the computer where the agent is installed. Users without this permission will see a numeric ID instead of the username.
* **View endpoint data that identifies the network:** Controls access to network-identifying data, such as private and public IP addresses, SSIDs, BSSIDs, gateway IP addresses, and gateway MAC addresses. Users without this permission will see a numeric ID instead of the network or device name.
* **View endpoint data that identifies location:** Controls access to location-related data, such as agent location, Wi-Fi location, and gateway location.
* **View endpoint data that identifies visited pages:** Controls access to Real User Test data, including visited websites, website IP addresses, and domains.

> **Note:** By default, all existing roles have these permissions enabled.

> **Note:** Agent names are derived from the computer name, which may contain the primary user's name. To avoid identifying an end-user through the agent name, administrators can either remove the relevant permission or update the agent name in the **Endpoint Agent > Agent Settings** page.

### Managing Endpoint PII Information

These permissions are designed to help organizations protect Personally Identifiable Information (PII) and remain compliant with privacy regulations such as the GDPR. When a user does not have one of these permissions, the system does not remove the data from the experience. Instead, it anonymizes the information.

* **Deterministic Anonymization:** The anonymization process is deterministic, meaning the same original value always produces the same anonymized value. This consistency allows users to correlate data across views, tables, and workflows even when sensitive details are hidden.
* **Collaborative Troubleshooting:** This approach supports collaboration between teams with different permission levels. A user without access to identifying information can share an anonymized agent name with a user who has full access. The user with the appropriate permissions can see both the original and anonymized values, allowing teams to collaborate on the same troubleshooting workflow while preserving privacy boundaries.

## Introduction to the Endpoint Agent Capabilities

### Scheduled Tests

Endpoint Agent and Pulse both run web-layer HTTP-server and network-layer Agent-to-Server **Scheduled Tests** types. Both test types are only IPv4 compatible. Scheduled tests can run at set intervals from a minimum of 1 minute to a maximum of 1 hour. All scheduled tests perform network-layer end-end and path trace tests. If a proxy is being used the network layer tests will target the proxy and not the end target. Scheduled Tests can be configured to run several types of checks against your target e.g. verify SSL, page content or HTTP response codes.

Endpoint Agent in conjunction with its web-browser plugin can be configured to monitor end-users web Browser Sessions automatically or end-users can manually start recording their **Browser Sessions** using their browser extension. When recording Browser Sessions the agent will also perform network-layer end-end and path trace tests and collect Network Access data. For full details on scheduled test configuration options see the [Creating Scheduled Tests on Endpoint Agents](https://docs.thousandeyes.com/product-documentation/endpoint-agent/creating-scheduled-tests-on-endpoint-agents) article.

### Dynamic Tests

Dynamic Tests (previously called Automated Session Tests) collect data from applications that are developed on Unified Communications as a Service (UCaaS) platforms; for example, Webex or Zoom. These tests are dynamic, so they execute only when a user uses an application on their device (therefore, the system and network resource utilization is optimal). The data collected from these tests can be used for troubleshooting issues faced by the user while using the application. Dynamic tests are a part of Synthetic tests and for instructions on how to configure and manage the Synthetic tests see the [Configuring options for Synthetic Tests](https://docs.thousandeyes.com/product-documentation/end-user-monitoring/test-settings/configure-tests) article.


---

# Agent Instructions: 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:

```
GET https://docs.thousandeyes.com/product-documentation/global-vantage-points/endpoint-agents.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
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.