No Result Found
Get your setup working faster. Join our Discord for optimisation tips from elite testers. Join our DiscordJoin our Discord

BrowserStack Load Testing with the MCP server

Use natural language prompts to create, run, and monitor load tests with the BrowserStack MCP server.

Use the BrowserStack MCP server to work with Load Testing directly from your AI assistant, such as Claude, Cursor, VS Code, GitHub Copilot, or ChatGPT. Create tests, start or stop runs, read reports, and check VU-hours with prompts. The assistant calls Load Testing with your credentials.

To set up the server, see get started with the BrowserStack MCP server.

Load Testing in the BrowserStack MCP server is being rolled out to accounts in phases. You need @browserstack/mcp-server v2.2.0 or later. If your account isn’t enabled yet, contact your BrowserStack account team to get access.

To use Load Testing tools on the MCP server, you need a BrowserStack account with Load Testing enabled for the MCP server. You also need an AI assistant that supports MCP, such as Claude, Cursor, VS Code, GitHub Copilot, or ChatGPT.

To create, edit, run, or stop load tests, you need the User or Admin role. The Viewer role is read-only.

Workflows and tools

Describe the outcome you want, and your AI assistant picks the right tool and runs it. You don’t need to name a tool in your prompt.

All test types: API (JMeter, k6, Gatling, Locust), Browser (Playwright, Selenium, WebdriverIO, Nightwatch), and Hybrid load tests.

Permissions apply to every workflow. The assistant asks you to confirm before it creates, edits, starts, or stops anything. For details, see Roles and permissions for Load Testing.

Projects

Ask your AI assistant to list your load testing projects, or to find a project or test by name. For example, “What load testing projects do I have?”

Workflow What it does Tool
List projects Returns your load testing projects, or finds one by name. listLoadTestProjects
List tests Returns the tests in a project, filtered by type, framework, or tag. listLoadTests
Find a test Finds a test by name across all projects. searchLoadTests
Get a test Returns the full configuration of a test. getLoadTest

Create and edit tests

Ask your AI assistant to create a test or change an existing one. For example, “Set Login Smoke to 200 VUs.”

Workflow What it does Tool
Create a test Creates an API, Browser, or Hybrid test from a sample script or your own script. createLoadTest
Update a test Changes a test’s name, tags, SLAs, script, or settings. updateLoadTest

When you create a test, the assistant asks for:

  1. Test type: API, Browser, or Hybrid.
  2. Framework: Choose based on test type, such as k6 for API or Playwright for Browser.
  3. Script: Sample or your own.
  4. Project: Existing or new.

The defaults are 1 virtual user, 30 seconds, constant load, and the default zone. You get a link to edit the test in the dashboard.

To use your own script, upload a zip file. The limit is 100 MB for API tests and 250 MB for Browser tests. Uploading your own script requires Claude Code, Cursor, or VS Code with local setup.

Run tests

Ask your AI assistant to estimate, start, check, or stop a run.

Workflow What it does Tool
Estimate cost Estimates the VU-hours that a run uses before you start it. estimateLoadTestRunCost
Start a run Starts a run of a test and returns the run ID, the estimated cost, and a report link. startLoadTestRun
Check run status Returns the live status and progress of a run. getLoadTestRunStatus
List active runs Returns the runs that are executing right now. listActiveLoadTestRuns
Stop a run Stops a running or queued run. Results collected before the stop are still available. stopLoadTestRun
List past runs Returns the past runs of a test, or all runs in a project. listLoadTestRuns
listProjectLoadTestRuns

Long runs continue in the background. To use different VUs or duration, ask to edit the test before you start the run.

Running a test uses VU-hours and sends real traffic to your target. Your assistant confirms the test and parameters first. To check the cost beforehand, ask your AI assistant to estimate the run’s VU-hours.

Read results

Ask about a completed run, such as “What was the error rate?” or “Which requests were slowest?”

Workflow What it does Tool
Get a run report Returns the full metrics for a run: KPIs, SLA results, transactions, and errors. getLoadTestRunReport
Get AI Insights Returns a summary, the root cause, and recommendations for a run. getLoadTestInsightSummary
List metrics Returns the metrics that are available for a test. getLoadTestMetricsManifest

Your assistant fetches the following from a run report:

  • Key metrics: p90, p95, p99 response time, request rate, and error rate.
  • Web Vitals for browser runs: LCP, INP, CLS, FCP, and TTFB.
  • Per-request stats and error breakdowns for 4xx and 5xx responses.
  • Threshold results.

For AI Insights, your assistant returns the same analysis as the Summary tab of the report. For more information, see AI Insights. If AI Insights are unavailable, you get the headline metrics instead.

Compare and trend

Ask your AI assistant to compare two runs, or to show how a metric changes over time. For example, “Did my latest run regress vs the previous one?” or “How has p95 changed over the last 10 runs?”

Workflow What it does Tool
Compare runs Compares two runs of the same test, with the baseline, new value, percentage change, and verdict for each metric. compareLoadTestRuns
Get test trends Returns the metric trends for one test over recent runs. getLoadTestHistoricalTrends
Get project trends Returns the metric trends across a project. getLoadTestProjectTrends

To compare runs, the following must be true:

  • The runs belong to the same test, and they’re different runs.
  • Both runs are completed.
  • Both runs have the same test type and framework.

The comparison also includes per-request changes and threshold results. For more information, see Compare reports.

VU-hours and quota

Ask your AI assistant how many VU-hours you have left, or how many a run will use. For example, “How many VU-hours will Login Smoke use?”

Workflow What it does Tool
Get quota Returns your included VU-hours, the VU-hours you used, the VU-hours you have left, and the parallel-run limit. getLoadTestQuota
Estimate cost Estimates the VU-hours for a planned run. estimateLoadTestRunCost

Your plan covers API and Browser tests. Browser VUs cost 10 times more than API VUs. When the load profile varies, the estimate is a range.

Platform status

Ask your AI assistant whether Load Testing is up. For example, “Is Load Testing working right now?”

Workflow What it does Tool
Check platform status Returns whether Load Testing is available. getLoadTestPlatformStatus

Use MCP with BrowserStack

Set up the BrowserStack MCP server once, and then use Load Testing from any supported AI assistant. Load Testing requires @browserstack/mcp-server v2.2.0 or later.

Set up the remote server

  1. Add the hosted MCP server URL https://mcp.browserstack.com/mcp to your assistant’s MCP configuration.
  2. Sign in with BrowserStack when prompted. No installation is required.

For Cursor and VS Code, you can use the one-click setup on the BrowserStack MCP server GitHub page.

Test your setup

Ask your AI assistant: “List my BrowserStack load testing projects.” If you get an error, see Troubleshoot the BrowserStack MCP server.

Limitations

These actions aren’t supported yet:

  • Delete tests or runs. Use the dashboard instead.
  • Schedule recurring runs. Use the dashboard or your CI/CD pipeline instead.
  • Compare different tests, or compare more than two runs. Ask for trends instead.

We're sorry to hear that. Please share your feedback so we can do better

Contact our Support team for immediate help while we work on improving our docs.

We're continuously improving our docs. We'd love to know what you liked





Thank you for your valuable feedback

Is this page helping you?

Yes
No

We're sorry to hear that. Please share your feedback so we can do better

Contact our Support team for immediate help while we work on improving our docs.

We're continuously improving our docs. We'd love to know what you liked





Thank you for your valuable feedback!

Talk to an Expert
Download Copy Check Circle