MCP Integration

PractiTest supports integration with AI assistants such as Claude using the Model Context Protocol (MCP). This integration connects your AI tool with PractiTest to create value-driven workflows, such as automatically generating tests from AI outputs, linking them to requirements, and adding them to PractiTest test sets for execution.

Important Notes: 

1. This capability is available only for Corporate licenses.
2. While the PractiTest MCP can be connected to different AI tools, it is tested and recommended for use with Claude.

Prerequisites

Before configuring the MCP, make sure you have the following:

• A PractiTest Corporate account with relevant permissions.
• The PractiTest MCP Server URL (see below).

Configuring MCP in your AI tool

Configuration Steps in Claude

1. Open your AI assistant and go to Settings.
2. Navigate to Connectors.
3. Click Add custom connector.
4. Enter a name for the connector.
5. In the Remote MCP Server URL field, paste the URL that matches your PractiTest data center:
   • US data center:
     https://prod.practitest.com/mcp/v1/server
   • EU data center:
     https://eu1-prod.practitest.app/mcp/v1/server
6. Save the configuration by clicking ‘Add’.
7. Then click on ‘Connect’ next to the connector, you’ll be redirected into PractiTest.
8. Click ‘Allow’ to complete the configuration.

Once completed, Claude will be connected to your PractiTest account and ready to use MCP actions.

Configuration Steps in ChatGPT

The following steps describe how to configure MCP using ChatGPT.

1. Open ChatGPT and go to Settings.
2. From the settings menu, select Apps and then open Advanced Settings.
3. Enable Developer Mode by toggling it on.
4. Once enabled, click Create App.
5. Enter a name for the app.
6. In the MCP Server URL field, paste the URL that matches your PractiTest data center:
   US data center:
   https://prod.practitest.com/mcp/v1/server
   EU data center:
   https://eu1-prod.practitest.app/mcp/v1/server
7. In the Authentication field, select OAuth.
8. Click Create to save the configuration.
9. You’ll be redirected into PractiTest, click ‘Allow’ to complete the configuration.

Once completed, ChatGPT will be connected to your PractiTest account and ready to use MCP actions.

Configuration Steps in GitHub Copilot (VS Code)

This guide walks you through how to connect the PractiTest MCP server to GitHub Copilot in VS Code, allowing you to interact with your PractiTest data directly from your development environment.

Before getting started, make sure you have:

• Visual Studio Code (latest version)
• GitHub Copilot extension installed
• GitHub Copilot Chat enabled

Step 1: Enable MCP in VS Code

1. Open Visual Studio Code
2. Go to Settings
   • Mac: Cmd + ,
   • Windows: Ctrl + ,
3. In the search bar, type: MCP
4. Under Chat settings, make sure:
   • Chat › MCP: Enabled is checked

Step 2: Open MCP Configuration in VS Code

1. Open VS Code
2. Open Command Palette: 
   • Mac: Cmd + Shift + P
   • Windows: Ctrl + Shift + P
3. Search for “MCP: Add Server” and select it

4. Select HTTP (HTTP or Server-Sent Events)

5. Paste paste the URL that matches your PractiTest data center: 

• US data center: https://prod.practitest.com/mcp/v1/server
• EU data center: https://eu1-prod.practitest.app/mcp/v1/server

7. Press “Enter” to confirm your input

8. Name your Connector and press “Enter” to confirm your input

Step 3: Verify Connection

1. Open Command Palette: 
   • Mac: Cmd + Shift + P
   • Windows: Ctrl + Shift + P
2. Search for “MCP: List Servers” 
3. You should see your server in the dropdown with a status “Running”. If the status is not “Running” click on the server and choose “Start Server”

Step 4: Test Connection

Test connection with a simple prompt in GitHub Copilot ChatExample prompt: Use My MCP server to list all my available projects in PractiTest

Available MCP Tools & How to Work With Them

PractiTest supports a set of MCP tools that allow AI tools to interact with your project data and perform actions directly in PractiTest. These tools can be used individually or combined to support complete workflows, from understanding coverage to creating tests.

For best results, keep your prompts clear and include the required context. We recommend starting each new conversation by specifying the project ID you want to work on, and referencing specific entity IDs (such as requirement ID or test ID) whenever relevant. You can find your project ID in the URL.

The table below describes the MCP tools currently available.

MCP Tool	Description	Example Prompt	Needed Input
Get Project list	Retrieves the list of projects available to the authenticated user, including project IDs. You can skip this by specifying your Desired project ID in the prompt.	“Show me the list of available PractiTest projects.” “Work on Project [project-id].”	Project ID (optional)
Get Requirement Coverage	Retrieves requirement coverage details including linked tests.	“Get coverage details for requirement [requirement-id].”	Requirement ID
Get Tests	Retrieves data for specific tests from PractiTest, including their details and steps.	“In Project [project-id], get details for test [test-id].”	Test ID, Project ID
Create a Test	Creates a new test in PractiTest in the selected project. Supports both Scripted and BDD tests.	Scripted: “In project [project-id], create a scripted test for login validation with steps for valid and invalid credentials.” BDD: “In project [project-id], create a BDD test for password reset, including scenarios for valid and invalid email.”	Project ID (or project context), test type (Scripted or BDD)
Link Test to a Requirement	Links a test to a requirement for traceability.	“Link the newly created tests to requirement [requirement-id].”	Test ID, Requirement ID
Create a Test Set	Creates a new test set in PractiTest within the specified project..	“In project [project-id], create a new test set called ‘Regression – Sprint 12’ for the upcoming release.”	Project ID (or project context), test set name, optional description
Add Test to a Test Set	Adds existing tests as instances to a specified test set.	“Add tests [Tests IDs] to test set [Test Set ID].”	Test Set ID, Test IDs (or test context)
Import Issues from Jira	Imports a Jira work item into PractiTest as an Issue using its Jira ID. Requires an active Jira integration and appropriate permissions for the selected project.	“Import Jira item [jira-item-id] into [practitest-project-id] as an Issue.”	Jira Work Item ID, PractiTest Project ID
Import Requirements from Jira	Imports a Jira work item into PractiTest as a Requirement using its Jira ID. Requires an active Jira integration and appropriate permissions for the selected project.	“Import Jira item [jira-item-id] into [practitest-project-id] as an Requirement.”	Jira Work Item ID, PractiTest Project ID

Usage Examples

Below are common examples of how PractiTest MCP tools can be used from your AI assistant. These examples demonstrate both simple test creation and a more advanced coverage analysis workflow.

1. Creating a Scripted Test

You can create a scripted test directly from your AI assistant by describing the test you want to generate.

Example prompt: Create a scripted test in project ID 12345 named “User Login Validation” with steps for successful login and invalid password handling.

This will create a new scripted test in the specified PractiTest project, including the test name, description, and steps.

2. Creating a BDD Test

You can also generate BDD tests with scenarios and steps.

Example prompt: Create a BDD test in project ID 12345 for password reset functionality, including scenarios for successful reset and invalid email.

The test will be created as a BDD test in PractiTest and will be available in the Test Library like any other BDD test.

3. Coverage Analysis and Test Generation Flow

MCP can be used to support an end-to-end coverage analysis workflow.

Example flow:

1. Specify the project you want to work on.
2. Request coverage information for a specific requirement.
3. Ask the AI assistant to analyze coverage gaps.
4. Instruct the assistant to create missing tests.
5. Link the newly created tests back to the requirement.

Example prompt: “In project ID 12345, analyze coverage for requirement ID 2, identify gaps, and suggest the missing tests.”

After reviewing and approving the suggestions, you can continue with:  “Create the tests in PractiTest and link them to requirement ID 2.”

This flow results in new tests created directly in PractiTest and fully linked to the original requirement, improving coverage and traceability.