Claude Cucumber Studio connectorWhat Claude can do in your Cucumber Studio account.
The Claude Cucumber Studio connector exposes 14 tools, and all 14 are read tools. 0 write, 0 undocumented. Below: what Claude can dig out of your test projects, how it installs on Claude Desktop, and who actually publishes it.
Verified Trustpilot reviews · AI, automation & growth agency
What it changes for a QA team
Instead of opening Cucumber Studio, filtering scenarios and digging through runs, you ask Claude what broke overnight or which tests cover checkout. It queries your account and answers in plain English. All 14 tools retrieve information; none of them changes it.
Check coverage in seconds. cucumberstudio_list_scenarios and cucumberstudio_find_scenarios_by_tags pull the scenarios of a project or a tagged scope, so "is this tested?" gets an answer without opening the app.
Make sense of a failure fast. cucumberstudio_get_test_run details a run and cucumberstudio_get_test_executions goes down to each individual result. You see which result failed, on which run.
Reuse before you rewrite. With cucumberstudio_list_action_words, Claude checks whether a shared step already exists before you author a new one, and cucumberstudio_get_action_word shows you its definition.
What Claude will not do here: by default, the sheet exposes reads only, so there is no tool to create a scenario, start a run or edit an action word. Nothing fires when a run finishes either. For that kind of chaining you need an automation tool, a different job covered on the Integrations hub.
The vocabulary in one minute
Five words before you install Cucumber Studio in Claude.
- Connector
- The link you set up once between Claude and an account you already have, so Claude can use it while answering you.
- Tool
- One named action a connector opens to Claude. Claude decides which ones to call; the directory lists them by name.
- Authorization
- The access you hand the service on Claude's behalf, granted once per person and yours to take back whenever you like.
- Approval
- The confirmation Claude waits for before an action that changes your account, shown in the chat right when it matters.
- MCP
- The shared standard behind connectors: it is what lets an assistant like Claude exchange data with an outside service.
Connect Cucumber Studio to Claude in three steps
- 01
Locate the connector
In Claude Desktop, open Customize, then Connectors, and find Cucumber Studio MCP. On a Team or Enterprise workspace, an Owner or Primary Owner enables it first, and each member then connects on their own.
- 02
Install the extension
The publisher's README describes an extension to install, not a sign-in page: download the latest .mcpb file from the repository's releases page, then import it into Claude Desktop.
- 03
Enter your API credentials
Open the extension's settings and fill in the three Cucumber Studio API values the README asks for: Access Token, Client ID and User ID. Claude then reaches what those credentials open, and nothing beyond.
The 14 tools, all read-only
Cucumber Studio MCP gives Claude 14 tools: 14 that read your account, 0 that change something in it.
Projects, scenarios, action words, runs, builds and environments: everything can be looked up, nothing can be changed. Tool names stay as Claude shows them.
- 14 read
What Claude reads (14)
14 toolsFourteen lookups, from the project list down to each test result.
cucumberstudio_list_projects
Lists every Cucumber Studio project your API credentials can reach, so the conversation starts from the right place instead of a guess at a project name.
cucumberstudio_get_project
Fetches the full detail of one specific project and reports it in the chat, so you do not have to click around Cucumber Studio.
cucumberstudio_list_scenarios
Returns the test scenarios stored in a given project. Claude gets an inventory of what already exists for that app, project by project.
cucumberstudio_get_scenario
Pulls the detailed information of one scenario into the conversation, so you can read what it covers without opening the app and clicking through the project.
cucumberstudio_find_scenarios_by_tags
Finds the scenarios that carry the tags you name. The README files this under scenario access: browsing test scenarios and searching them by tag, rather than combing through a whole project.
cucumberstudio_list_action_words
Lists the reusable action words of a project. In Cucumber Studio, these are the test steps and definitions that several scenarios share, so the list shows what already exists.
cucumberstudio_get_action_word
Opens the detailed information of one action word, the shared step your scenarios reuse, so you can reread how it is defined from inside the chat.
cucumberstudio_find_action_words_by_tags
Narrows the action words down to those with certain tags, useful once the shared step library has grown large.
cucumberstudio_list_test_runs
Goes through the test runs recorded in Cucumber Studio. Runs belong to the test execution side of the tool, which the README describes as viewable, never as something the server launches.
cucumberstudio_get_test_run
Gives the detailed information of one particular test run, straight in the chat, once you know which run you are after from the list.
cucumberstudio_get_test_executions
Returns the individual test results, execution by execution. That is the level of detail you need to spot what failed inside a run, rather than reading a single global status.
cucumberstudio_list_builds
Lists the builds recorded in Cucumber Studio. Build information is part of the test execution data the README says the server lets you view.
cucumberstudio_get_build
Shows the detailed information of a single build, picked from the list returned by the build listing tool, without leaving the conversation.
cucumberstudio_list_execution_environments
Lists the execution environments set up in your Cucumber Studio project, the last of the fourteen lookups the README publishes.
What Claude asks before it acts
By default, Claude stops and asks for your go-ahead before each action it takes on an account for you. Here, all 14 tools only read.
No source describes a confirmation specific to this connector. On a Team or Enterprise workspace, owners decide whether a member may let certain actions through without being asked again. They can also cap what a connector is allowed to do across the organization, keeping reads and blocking writes for example, and nobody gets around that from their own account. Claude works with your credentials and nothing more: a project you cannot reach stays closed to it.
Which plans it is available on
None of the 819 sheets in the directory shows plan availability. That information is not published anywhere, connector by connector.
The general rule is published: remote connectors are open to all users on Claude, Cowork, Claude Desktop and mobile, and desktop extensions like this one install on Claude Desktop. On Team and Enterprise, an Owner or Primary Owner turns the connector on for the organization first. For the current state on your account, the connector's sheet in the official directory is the only reference.
Where this connector stops
A connector is not an automation. Claude calls these tools while it answers you, and nothing starts on its own when a test run ends.
Only the publisher documents this connector, through the README of its repository: neither Claude's help center nor Anthropic's developer docs cover it. The tool list is an observed floor, not a promise. The directory badge is not a security audit either, and Anthropic says so on every sheet: it does not choose the tools a publisher exposes and does not guarantee how they behave. Check the repository before you hand over your credentials.
Need help connecting Cucumber Studio MCP to Claude?
A person reads every message.
