MCP test automation
MCP test automation: let AI agents and your team drive one test platform
Claude Code or any MCP client creates, runs and analyzes your web, mobile and API tests, in the repository your team reviews.
See the 152 MCP toolsOff by default · in every plan
What is MCP test automation?
Definition
MCP test automation means letting an AI agent create, run and analyze automated tests through tools it calls over the Model Context Protocol (MCP), an open standard. The Cerberus Testing MCP server connects the agent to your test platform.
Tests that land in your repository
What the agent writes becomes a test case your team can open, not a chat.
Verdicts from actual runs
Verdicts and evidence come from an actual execution, never from a model’s guess.
An open standard, no lock-in
Claude Code or any MCP client connects; Cerberus Testing has been open source since 2013.
- Anthropic introduces the Model Context Protocol as an open standard. Source (opens in a new tab)
- MCP joins the Agentic AI Foundation, under the Linux Foundation. Source (opens in a new tab)
Two kinds of MCP servers for testing
An MCP server for testing either drives a browser or works on the tests themselves. Cerberus Testing’s belongs to the second family.
Browser MCP servers: Playwright MCP, Chrome DevTools MCP
They give an agent a browser.
- The agent can
- Open a page, read its accessibility tree, click, type, take screenshots.
- Playwright MCP tools
browser_navigate browser_snapshot browser_click browser_type
- What it produces
- Actions in a live browser, and Playwright code for them (
--codegen). - Best for
- Exploring a page, reproducing a bug.
Test-platform MCP server: Cerberus Testing
It gives an agent your test platform.
- The agent can
- Create, run and analyze tests, campaigns, schedules, robots and data: 152 tools.
- Cerberus Testing tools
cerberus_testcase_ create cerberus_campaign_ execution_ create cerberus_tag_ wait cerberus_testcase_ execution_ get
- What it produces
- Test cases, runs and results your whole team opens in the web interface.
- Best for
- A regression suite the whole team runs, across web, mobile and APIs.
Playwright MCP gives an agent a browser. Cerberus Testing gives it your whole test platform.
The 152 tools of the Cerberus Testing MCP server
From test folders to robots: 152 tools in 17 areas, each with a single job.
152MCP tools in 17 areas
- 66 read
- 28 create
- 2 launch a test or a campaign
- 29 update, move or reorder
- 24 delete or detach
- 3 pause, resume or cancel a run
24 tools declare it: 23 deletions and cerberus_tag_cancel. Your MCP client can ask for your approval before running one.
Execution
Resolve where a test can run, launch it, read the result and its evidence.
cerberus_The country, environment and robot combinations that can really run a test case.testcase_ execution_ targets cerberus_Queues and triggers a test case on one or more countries and environments; returns a tag.testcase_ execution_ create cerberus_Runs a whole campaign, with optional overrides; returns a tag.campaign_ execution_ create cerberus_Executions and queue entries, by tag or by test case.testcase_ execution_ list cerberus_Result of an execution: status, and where and why it failed.testcase_ execution_ get cerberus_An execution artifact: page source, service response, log, or the screenshot as an image.testcase_ execution_ file_ get
Run control
Wait for a run by its tag and read the verdict, or pause, resume and cancel it.
cerberus_Waits for a run to finish, up to 55 s per call, then returns counters, CI verdict and failures.tag_ wait cerberus_Aggregated result of a tag: CI score, flaky tests.tag_ get cerberus_Pauses what has not started yet.tag_ pause cerberus_Resumes a paused run.tag_ resume cerberus_Cancels what has not started; running tests finish normally.tag_ cancel
Test cases
Create, read and search test cases, including the sequence that actually runs.
cerberus_Lists the test cases of a folder.testcase_ list cerberus_Details of a test case, with its steps if asked.testcase_ get cerberus_The sequence that actually runs, library steps resolved.testcase_ scenario_ get cerberus_Every place a string appears: selector, URL, value, property.testcase_ search_ content cerberus_Creates a test case in a folder; the ID is assigned for you.testcase_ create cerberus_Updates, activates or deactivates a test case.testcase_ update cerberus_Deletes a test case.testcase_ delete
Steps
Steps of a test case, and reusable library steps.
cerberus_Steps of a test case, as stored.testcase_ step_ list cerberus_Details of one step.testcase_ step_ get cerberus_Adds a step, at a given position if needed.testcase_ step_ create cerberus_Updates a step.testcase_ step_ update cerberus_Deletes a step with its actions and controls.testcase_ step_ delete cerberus_Copies a step, actions and controls included.testcase_ step_ duplicate cerberus_Sets the order in which steps run.testcase_ step_ reorder cerberus_Reusable library steps of a system.step_ library_ list cerberus_Adds a step that points to a library step: fix it once, and it’s fixed everywhere.testcase_ step_ library_ use
Actions
What each step does: open a URL, click, type, call a service…
cerberus_Actions of a step.testcase_ step_ action_ list cerberus_Details of one action.testcase_ step_ action_ get cerberus_Adds an action: openUrl, click, type, callService…testcase_ step_ action_ create cerberus_Updates an action.testcase_ step_ action_ update cerberus_Deletes an action.testcase_ step_ action_ delete cerberus_Moves an action, with its controls, to another step.testcase_ step_ action_ move cerberus_Sets the order in which the actions of a step run.testcase_ step_ action_ reorder
Controls
The assertions that decide whether an action passed.
cerberus_Controls (assertions) of an action.testcase_ step_ action_ control_ list cerberus_Details of one control.testcase_ step_ action_ control_ get cerberus_Adds a control, such as verifyElementPresent.testcase_ step_ action_ control_ create cerberus_Updates a control.testcase_ step_ action_ control_ update cerberus_Deletes a control.testcase_ step_ action_ control_ delete cerberus_Moves a control to another action.testcase_ step_ action_ control_ move cerberus_Sets the order in which the controls of an action run.testcase_ step_ action_ control_ reorder
Countries and properties
Where a test case can run, and the %property.NAME% values it uses per country.
cerberus_Countries a test case can run in.testcase_ country_ list cerberus_Checks one country of a test case.testcase_ country_ get cerberus_Adds a country to a test case.testcase_ country_ create cerberus_Removes a country from a test case.testcase_ country_ delete cerberus_Properties a test case can use, written %property.NAME%.testcase_ country_ property_ list cerberus_One property, with its countries.testcase_ country_ property_ get cerberus_Creates a property, for one country or all.testcase_ country_ property_ create cerberus_Updates a property for a country.testcase_ country_ property_ update cerberus_Deletes a property for a country.testcase_ country_ property_ delete
Test folders
The folders that organize the test repository.
cerberus_Lists the test folders.test_ folders_ list cerberus_Details of one test folder.test_ folder_ get cerberus_Creates a test folder.test_ folder_ create cerberus_Updates, activates or deactivates a test folder.test_ folder_ update cerberus_Deletes a test folder.test_ folder_ delete
Labels
Test case labels, system labels, and the labels that select a campaign’s tests.
cerberus_Labels attached to a test case.testcase_ label_ list cerberus_Labels a test case, which puts it in the campaigns using that label.testcase_ label_ add cerberus_Detaches labels from a test case.testcase_ label_ remove cerberus_Labels defined in a system.system_ label_ list cerberus_Defines a new label in a system.system_ label_ create cerberus_Labels that select a campaign’s test cases.label_ list cerberus_One label attached to a campaign.label_ get cerberus_Attaches a label to a campaign.label_ create cerberus_Changes which label a campaign attachment points to.label_ update cerberus_Removes a label from a campaign.label_ delete
Campaigns
Campaigns, their criteria, cron schedules and notifications.
cerberus_Lists campaigns.campaign_ list cerberus_A campaign’s settings, labels, countries, environments and robots.campaign_ get cerberus_Creates a campaign: a reusable set of tests and where to run them.campaign_ create cerberus_Updates a campaign.campaign_ update cerberus_Deletes a campaign.campaign_ delete cerberus_Countries, environments, robots and filters of a campaign.campaign_ parameter_ list cerberus_One campaign criterion.campaign_ parameter_ get cerberus_Adds a criterion: country, environment, robot, status, priority…campaign_ parameter_ create cerberus_Changes a campaign criterion in place.campaign_ parameter_ update cerberus_Removes a campaign criterion.campaign_ parameter_ delete cerberus_Cron schedules of a campaign.campaign_ scheduler_ list cerberus_One schedule: cron, active flag, last run.campaign_ scheduler_ get cerberus_Schedules a campaign with a Quartz cron expression.campaign_ scheduler_ create cerberus_Updates a schedule.campaign_ scheduler_ update cerberus_Removes a schedule.campaign_ scheduler_ delete cerberus_Notifications fired when a campaign starts or ends.eventhook_ list cerberus_One notification hook.eventhook_ get cerberus_Notifies Slack, Teams, email, Google Chat or a webhook, optionally only on failure.eventhook_ create cerberus_Updates a notification hook.eventhook_ update cerberus_Removes a notification hook.eventhook_ delete
Applications and objects
Applications under test and their reusable locators.
cerberus_Applications under test.application_ list cerberus_Details of an application.application_ get cerberus_Creates an application: name, type, system.application_ create cerberus_Updates an application.application_ update cerberus_Deletes an application.application_ delete cerberus_Reusable locators (XPath, CSS…) of an application.application_ object_ list cerberus_One application object.application_ object_ get cerberus_Creates a locator that actions reference by name.application_ object_ create cerberus_Updates a locator once, for every test that uses it.application_ object_ update cerberus_Deletes an application object.application_ object_ delete
Services
REST, SOAP, Kafka, MongoDB and FTP calls, with headers and content.
cerberus_Services Cerberus can call.appservice_ list cerberus_A service with its headers and content.appservice_ get cerberus_Creates a service, straight from a whole curl command.appservice_ create cerberus_Updates a service.appservice_ update cerberus_Deletes a service.appservice_ delete cerberus_HTTP headers of a service.appservice_ header_ list cerberus_One header of a service.appservice_ header_ get cerberus_Adds an HTTP header to a service.appservice_ header_ create cerberus_Updates a header.appservice_ header_ update cerberus_Removes a header.appservice_ header_ delete cerberus_Content parameters: form data, URL-encoded.appservice_ content_ list cerberus_One content parameter.appservice_ content_ get cerberus_Adds a content parameter.appservice_ content_ create cerberus_Updates a content parameter.appservice_ content_ update cerberus_Removes a content parameter.appservice_ content_ delete
Data library
Test data from SQL, services, CSV files or internal values.
cerberus_Data libraries a getFromDataLib property can use.datalib_ list cerberus_A data library, its entries and column mapping.datalib_ get cerberus_Creates a library and its entries in one call: SQL, service, CSV, internal.datalib_ create cerberus_Corrects a data library’s definition or mapping.datalib_ update
Environments
Country and environment pairs, and how each application is reached.
cerberus_Country and environment pairs enabled for a system.country_ env_ param_ list cerberus_One environment: build, revision, maintenance.country_ env_ param_ get cerberus_Enables a country and environment pair (DEV, QA, PROD…).country_ env_ param_ create cerberus_Updates an environment configuration.country_ env_ param_ update cerberus_Removes an environment configuration.country_ env_ param_ delete cerberus_How each application is reached, per environment.country_ environment_ parameters_ list cerberus_Endpoint, context root, login URL and pool size of one application.country_ environment_ parameters_ get cerberus_Declares how an application is reached on an environment.country_ environment_ parameters_ create cerberus_Updates how an application is reached.country_ environment_ parameters_ update cerberus_Removes how an application is reached.country_ environment_ parameters_ delete
Robots and executors
Execution profiles, WebDriver capabilities and the hosts that run them.
cerberus_Execution profiles: platform, browser, version.robot_ list cerberus_A robot with its capabilities and executors.robot_ get cerberus_Creates a robot.robot_ create cerberus_Updates a robot.robot_ update cerberus_Deletes a robot.robot_ delete cerberus_WebDriver capabilities of a robot.robot_ capability_ list cerberus_One capability.robot_ capability_ get cerberus_Adds a WebDriver capability (key and value).robot_ capability_ create cerberus_Updates a capability.robot_ capability_ update cerberus_Removes a capability.robot_ capability_ delete cerberus_Hosts able to run a robot.robot_ executor_ list cerberus_One executor: host, port, proxy.robot_ executor_ get cerberus_Adds a host that runs a robot.robot_ executor_ create cerberus_Updates an executor.robot_ executor_ update cerberus_Removes an executor.robot_ executor_ delete
Invariants
Reference lists: countries, environments, browsers, priorities, statuses…
cerberus_Values of a reference list, for one type (countries, environments…).invariant_ list cerberus_One reference value.invariant_ get cerberus_Adds a reference value.invariant_ create cerberus_Updates a reference value.invariant_ update cerberus_Deletes a reference value.invariant_ delete
Context and reference
Active workspaces, Cerberus variables, and the catalogs of actions, controls, conditions and property types.
cerberus_Your active workspaces, and the ones you are allowed to activate.context_ system_ list cerberus_Adds or removes workspaces from your active context.context_ system_ update cerberus_Variables Cerberus substitutes in a value, and when each one resolves.variable_ list list_Actions available for an application type.available_ actions list_Controls (assertions) available for an application type.available_ controls list_Condition operators that decide whether a step or action runs.available_ conditions list_Property types: text, SQL, JSONPath, OTP…available_ property_ types get_Version of the Cerberus database schema.db_ version
Names as your MCP client lists them. Source: cerberustesting/cerberus-core (opens in a new tab).
Cerberus Testing and Playwright: complementary
Playwright suits tests versioned with your code; Cerberus Testing suits campaigns shared with QA and business teams.
| Criterion | Playwright with Playwright MCP and Test Agents | |
|---|---|---|
| What it is | A library and test runner; Playwright MCP drives a browser. | A test automation platform with a built‑in MCP server. |
| Who writes tests | Developers and SDETs, comfortable with code. | The whole team, in a low‑code editor, and AI agents through MCP.Strength |
| What its MCP server gives an agent | A browser: navigate, read the accessibility tree, click, type. | Test repository, runs, campaigns, robots and data: 152 tools.Strength |
| Scope | Web (Chromium, Firefox, WebKit), API, mobile emulation; Android is experimental. | Web (Selenium), native Android and iOS (Appium), desktop (Sikuli), REST, SOAP, Kafka, MongoDB, FTP, SQL, batch.Strength |
| Where tests live | Files in your Git repository, with code review and diffs.Strength | A central repository shared by the team; versioning differs from Git. |
| Reports and debugging | HTML report, Trace Viewer, UI mode.Strength | Execution and campaign reports; screenshots, video, logs, network traffic. |
| Running at scale | Runner parallelism and sharding, in CI.Strength | Campaigns your team launches or schedules from the web interface: a queue, parallel runs across countries and environments, a CI score.Strength |
| Built‑in AI | Test Agents (planner, generator, healer) for test code. | AI Assistant (Enterprise extra) that calls the same 152 tools, with your Anthropic key. |
| License | Apache-2.0, maintained by Microsoft. | GPLv3, open source since 2013. |
| Hosting | Runs locally or in CI; no server. | Cerberus Testing Cloud, managed for you, from a free plan to Enterprise. |
| Community | Very large: about 97,000 GitHub stars.Strength | Smaller; in production at La Redoute and MyTF1. |
Playwright facts: playwright.dev (opens in a new tab) and GitHub, September 2026.
Get started: connect Claude Code in 4 steps
A cloud instance, one setting, one key and one command.
Start with Cerberus Testing Cloud
The free plan gives one user a dedicated instance; the MCP server is included in every plan.
Turn on the server, get your API key
An administrator sets
cerberus_mcp_enabletotruein Admin › Parameters, then creates your key in Admin › Users.Admin › ParametersValue
cerberus_mcp_ enable Boolean in order to enable the MCP endpoints.
falsetrueAdmin › UsersYour user
API Key
Set New KeyCopyAdd the server to Claude Code
One command, with your instance’s
/mcpaddress and your key in theX-API-KEYheader.claude mcp add --transport http cerberus https://your-cerberus/mcp --header "X-API-KEY: <your-key>"Prefer JSON? Same settings for other MCP clients with remote HTTP support
{ "mcpServers": { "cerberus": { "type": "http", "url": "https://your-cerberus/mcp", "headers": { "X-API-KEY": "<your-key>" } } } }Ask, then read the result
The agent resolves the targets, runs the test, waits with
cerberus_tag_waitand reports the verdict.“Run the login test on QA and tell me the verdict.”
What each Cerberus Testing Cloud plan includes
The MCP server comes with every plan; the built‑in AI Assistant is an Enterprise extra.
/mcp, off by defaultIncluded in Free planIncluded in Starter · BusinessIncluded in EnterpriseFAQ
MCP test automation: FAQ
What is MCP in test automation?
What can an AI agent do with the Cerberus Testing MCP server?
What is the difference between Playwright MCP and the Cerberus Testing MCP server?
Which AI clients can connect?
X-API-KEY header or uses OAuth 2.0 through Keycloak. The MCP Inspector gives you the configuration to copy.Is the MCP server included in my plan?
cerberus_mcp_enable.How much does the AI cost?
Is it safe to let an AI agent run my tests?
Where does my test data go?
Do I need to code to automate tests with AI?
Will AI agents replace QA testers?
Get started
Give your agents a test platform, not just a browser
Start MCP test automation with Cerberus Testing Cloud, for free. Your agents get the MCP server; your team gets the web interface.
Play with the demo(opens the demo login in a new tab)
Review the main use‑cases right away.
Talk to our team
Free demo, needs assessment, any question.

