Connect to NeoLoad Web MCP

The NeoLoad Web Model Context Protocol (MCP) integration lets you interact with your performance testing platform through natural language conversations with AI assistants. Instead of navigating complex UIs or writing scripts, you can manage tests, analyze results, and generate reports through simple conversational prompts.

MCP transforms your performance testing experience by enabling you to:

Discover and manage workspaces and tests through natural language.
Execute performance tests with conversational commands.
Analyze results and generate insights automatically.
Monitor test execution in real-time.
Create comprehensive reports without manual data gathering.

MCP requires a compatible AI assistant that supports the Model Context Protocol standard. Currently tested and supported platforms include Cursor and Claude Desktop.

Before you start

Before you connect to NeoLoad MCP, make sure you have:

An API token from NeoLoad Web.
A compatible AI assistant (Cursor or Claude Desktop).
Node.js v18 or later installed on your system.

Get your API token

To connect to the MCP server, you need a valid API token from NeoLoad Web. To get your API token, follow these steps:

Sign in to NeoLoad Web.
Select User menu > Profile.
Go to the Access tokens section.
Select + Generate token.
Copy and securely save your token.

Configure your AI assistant

Configure your AI assistant to connect to the NeoLoad Web MCP server. The configuration steps depend on which AI assistant you're using.

Cursor configuration

To configure Cursor to connect to NeoLoad Web MCP, follow these steps:

Open your Cursor settings and open the Tools & Integrations tab.
Select New MCP Server.

Add the NeoLoad Web MCP server configuration using the appropriate server URL for your region:

US Region: https://neoload-v2.saas.neotys.com/mcp-server/mcp
EU Region: https://neoload-v2-eu.saas.neotys.com/mcp-server/mcp

Copy

{
  "mcpServers": {
    "nlweb-mcp-server": {
      "command": "npx",
      "args": [
        "mcp-remote",
        "https://neoload-v2.saas.neotys.com/mcp-server/mcp",
        "-y",
        "--header",
        "Authorization:${AUTH_HEADER}"
      ],
      "env": {
        "AUTH_HEADER": "<your NLW API token>"
      }
    }
  }
}

Enter your API token in the authentication settings.

Keep the server name short. Cursor limits the combined MCP server name and tool name to 60 characters.

Claude Desktop configuration

To configure Claude Desktop to connect to NeoLoad Web MCP, follow these steps:

Open your Claude Desktop configuration file.

Add the NeoLoad Web MCP server configuration using the appropriate server URL for your region:

US Region: https://neoload-v2.saas.neotys.com/mcp-server/mcp
EU Region: https://neoload-v2-eu.saas.neotys.com/mcp-server/mcp

Copy

{
  "mcpServers": {
    "nlweb-mcp-server": {
      "command": "npx",
      "args": [
        "mcp-remote",
        "https://neoload-v2.saas.neotys.com/mcp-server/mcp",
        "-y",
        "--header",
        "Authorization:${AUTH_HEADER}"
      ],
      "env": {
        "AUTH_HEADER": "<your NLW API token>"
      }
    }
  }
}

Replace <your NLW API token> with your actual API token.

Verify your connection

To verify that your MCP connection is working correctly, follow these steps:

Open your AI assistant.
Start a conversation with a simple command like: "Show me all available workspaces"
If the connection is successful, you should see a list of your NeoLoad Web workspaces.

If you encounter connection issues, verify that:

Your API token is valid and correctly entered.
You're using the correct server URL for your region.
The MCP feature is enabled for your account.
Node.js v18 or later, and required dependencies are properly installed.

Available MCP actions

Once connected, you can use these MCP actions to manage your performance testing workflow through natural language conversations:

Workspace and test management

Manage your workspaces and run performance tests with these actions:

Action	Description	Parameters
list-workspaces	Get a list of all workspaces you can access.	None
list-tests-workspace	Get all tests in a workspace.	workspaceId, pageNumber, pageSize, sort
list-results-workspace	Get test results from a workspace.	workspaceId, pageNumber, pageSize, sort
run-test	Start a performance test.	testId, resultName
stop-test-execution	Stop a test that's currently running.	testExecutionId
get-test-project	Get test available scenarios.	None
get-test-scenario-populations	Get populations from a specific test.	testId
update-test-scenario-populations	Update populations in a test scenario.	testId, populations, constant_load, rampup_load, peaks_load, custom_load
update-selected-test-scenario	Update the selected scenario in a test.	None
update-test-zone-configuration	Reads all populations from the scenario, suggests optimal zone distribution, considers population-specific requirements, recommends zone sizing per population, and updates the test.	testId, controllerZoneId (optional), lgsByZone (optional), cloudLgResourcesSize (optional)

Result analysis

Analyze and examine your test results in detail with these actions:

Action	Description	Parameters
get-result	Get detailed information about a test result.	resultId
get-result-elements	Get the elements from a test result.	resultId
get-result-events	Get events that occurred during test execution.	resultId, pageNumber, pageSize, sort
get-result-logs	Get logs from test execution.	resultId, sort
get-result-requests-values	Get performance metrics for requests.	resultId, intervalId (optional), pageNumber, pageSize, sort
get-result-transactions-values	Get performance metrics for transactions.	resultId, intervalId (optional), pageNumber, pageSize, sort
get-result-timeseries	Get timeseries metrics for a test result (user load, errors, response time, throughput) as sampled data points. Supports completed results and live polling for running results.	resultId, fixedWindowDuration (optional), intervalId (optional), requestToken (optional)
get-result-contexts	Get the valid filter values for populations, user paths, and zones available for a test result. Provides the context needed before calling other filtered analysis actions.	resultId
get-result-slas	Get the Service Level Agreement (SLA) evaluation results for a test result, including pass, warn, or fail status and threshold details. Accepts filters for SLA type, element type, or status.	resultId, slaType, elementType (optional), slaStatus (optional), pageNumber (optional), pageSize (optional), sort (optional)
get-result-statistics	Get the full summary statistics for a test result in one call, including request, transaction, and iteration totals. Includes live values while a test is running.	resultId

Result management

Update test result metadata after analysis with these actions:

Action	Description	Parameters
update-result	Update a test result's name, description, lock status, or quality status. Use this after analysis to write findings, such as `PASSED` or `FAILED`, to the result, or to lock the result to prevent automatic deletion.	resultId, name (optional), description (optional), lock (optional), qualityStatus (optional)

Interval management

Create and manage time intervals for detailed test result analysis with these actions:

Action	Description	Parameters
generate-intervals	Create intervals to analyze test results.	resultId
list-intervals	Get all intervals from a test result.	resultId

Result monitors

Access and analyze monitor metrics from your test results with these actions:

Action	Description	Parameters
list-result-monitors	Get all available monitor elements for a test result.	resultId
get-result-monitors-values	Get current or aggregated values for selected monitor elements.	resultId, monitorIds (optional), pageNumber, pageSize, sort
get-result-monitors-timeseries	Get time series data for selected monitors over configurable time windows.	resultId, monitorIds, startTime (optional), endTime (optional)

Zone management

Configure and manage your load testing zones with these actions:

Action	Description	Parameters
list-zones	Get available zones, optionally filtered by type.	type (STATIC, DYNAMIC, CLOUD)
create-static-zone	Create a static zone.	zoneName
create-dynamic-zone	Create a dynamic zone with specified resources.	zoneName, providerId, sizing
enable-cloud-zone	Enable a cloud zone in a specific region.	cloudRegion
disable-cloud-zone	Turn off a cloud zone.	zoneId
update-dynamic-zone-resource-sizing	Change resource sizing for a dynamic zone.	zoneId, sizing
rename-static-zone	Change the name of a static zone.	zoneId or name, newName
delete-static-dynamic-zone	Remove a static or dynamic zone.	zoneId or name

Dashboard management

Create and manage dashboards with AI-powered insights using these actions:

Action	Description	Parameters
list-dashboards	Get all available dashboards with filtering and sorting.	workspaceId (optional), testResultId (optional), sort (optional), pageNumber (optional), pageSize (optional)
create-dashboard	Create a new dashboard in a workspace.	workspaceId, name, dashboardType (optional)
get-dashboard-tiles	Get all tiles and their configuration from a dashboard.	dashboardId
create-dashboard-tile	Add a new tile to a dashboard with position and size.	dashboardId, title, x, y, width (optional, default: 1), height (optional, default: 1)
add-data-series-to-dashboard-tile	Add a data series to a tile, such as element, time series, or monitor data.	dashboardId, tileId, dataSeries (includes filter, color, legend, resultId, resultName, visible)

Infrastructure management

Monitor and check your infrastructure providers with these actions:

Action	Description	Parameters
list-infrastructure-providers	Get all available infrastructure providers.	None
check-infra-provider-availability	Check if an infrastructure provider is available.	providerId

Webhook management

Access and manage webhook configurations for your workspaces with these actions:

Action	Description	Parameters
get-webhooks	Get webhooks configured for a workspace.	workspaceId

What's next?

Now that you've connected to NeoLoad Web MCP, you can:

Learn MCP prompting best practices to get the most out of your natural language interactions.