Zephyr Scale

This MCP server acts as a digital bridge that allows AI assistants to talk directly to Zephyr Scale, a popular tool for managing software tests inside Jira. Instead of a human manually typing out test cases or clicking through Jira menus, this tool lets an AI model handle the heavy lifting of organizing and documenting quality assurance tasks. It makes it easy for developers to automate the creation of test scenarios, search for existing tests, and keep their testing workflow organized without ever leaving their AI-powered workspace. Under the hood, the server provides a comprehensive suite of tools for managing the entire test lifecycle, from initial creation and folder organization to test runs and execution tracking. It is designed to be versatile, supporting both Jira Cloud and Jira Data Center by automatically detecting the environment to ensure it uses the correct Atlassian REST API version. A particularly clever feature is its unified resource system, which allows an AI to pull live test cases from a Jira instance to use as templates via a custom URI scheme. This ensures that any new content generated by the AI remains consistent with existing project standards and specific custom fields. For developers integrating LLMs into their development environments, this server is especially powerful because it uses official API-compliant schemas. This structure helps the AI understand the exact data requirements for various test formats—including BDD (Gherkin), step-by-step, and plain text scripts—which significantly reduces formatting errors. By exposing these capabilities through the Model Context Protocol, the tool transforms a static Jira instance into a dynamic, context-aware database that an AI can query, update, and expand upon in real-time, greatly accelerating the path from initial requirements to verifiable test results.

Category: Developer Tools & Code Intelligence

Tags: atlassian, jira, qa, testing, zephyr scale

Visit Zephyr Scale

How to install and configure Zephyr Scale

1. Installation You can run the Zephyr Scale MCP server using npx (which requires no permanent installation) or install it globally via npm. Option 1: npx (Recommended) No installation command is required; you simply reference the package in your configuration. Option 2: Global Installation bash npm install -g zephyr-scale-mcp-server ---

2. Configuration Add the following to your MCP settings file (e.g., claude_desktop_config.json). The server automatically detects your environment based on the ZEPHYR_BASE_URL, but you can explicitly set JIRA_TYPE to "cloud" or "datacenter" in the environment variables if needed.

For Jira Cloud json { "mcpServers": { "zephyr-server": { "command": "npx", "args": ["zephyr-scale-mcp-server@latest"], "env": { "ZEPHYR_BASE_URL": "https://your-company.atlassian.net", "JIRA_USERNAME": "your-email@example.com", "JIRA_API_TOKEN": "your-api-token" } } } }

For Jira Data Center json { "mcpServers": { "zephyr-server": { "command": "npx", "args": ["zephyr-scale-mcp-server@latest"], "env": { "ZEPHYR_BASE_URL": "https://your-jira-server.com", "ZEPHYR_API_KEY": "your-api-token" } } } } Note: If you performed a global npm installation, change "command": "npx" and "args": ["zephyr-scale-mcp-server@latest"] to "command": "zephyr-scale-mcp". ---

3. Available Tools The server provides tools for managing the full test lifecycle: Test Case Management * get_test_case: Retrieve detailed information about a specific test case. * create_test_case: Create test cases (supports STEP_BY_STEP, PLAIN_TEXT, or BDD content). * update_test_case_bdd: Update an existing test case with BDD content. * delete_test_case: Remove a specific test case. Test Run Management * create_test_run: Initialize a new test run. * get_test_run: Get detailed information about a specific test run. * get_test_run_cases: Retrieve test case keys associated with a test run. * add_test_cases_to_run: Add test cases to an existing test run. Test Execution & Search * get_test_execution: Retrieve specific individual test execution results. * search_test_cases_by_folder: Locate test cases within a specific folder. Organization * create_folder: Create a new folder within Zephyr Scale. Resources * zephyr://testcase/YOUR-TEST-CASE-KEY: Fetch real test case data to use as templates. * file:///path/to/file.json: Access local files. * zephyr://examples/...: Access built-in example payloads. ---

4. Example Prompts Create a BDD Test Case "Create a new BDD test case in project 'PROJ' named 'User Authentication'. The script should cover a valid user login scenario: Given a user with valid credentials, when the user attempts to log in, then the user should be authenticated successfully." Create a Test Run "Create a new test run for project 'PROJ' titled 'Sprint 1 Test Run' and include test cases PROJ-T123, PROJ-T124, and PROJ-T125." Template-based Creation "Fetch the test case PROJ-T123 to use as a template. Using its folder and custom field structure, create a new test case for 'Password Reset' in the same project." Search and Organize "Find all test cases in the 'Regression' folder and add them to a new test run called 'Weekly Regression'."

What you can do with Zephyr Scale

Use Case 1: Automated Test Generation from Requirements Problem: Manual translation of Product Requirements Documents (PRDs) or user stories into structured test cases is a tedious, error-prone process that often delays the start of the testing cycle. Solution: By combining an AI agent with the Zephyr Scale MCP, teams can automatically generate full test suites directly from requirements. The AI analyzes the feature description, identifies edge cases, and uses the create_test_case tool to populate Jira with BDD or Step-by-Step instructions. Example: A developer provides a PRD for a "Password Reset" feature. The AI analyzes the requirements and immediately creates five BDD test cases in Zephyr Scale covering successful resets, expired tokens, and mismatched passwords, placing them in the correct "Security" folder.

Use Case 2: Maintaining Consistency via Live Templating Problem: Large QA teams often struggle with inconsistent test case metadata. New test cases frequently miss mandatory custom fields, labels, or specific project configurations, making reporting difficult. Solution: This MCP uses a unified resource system (`zephyr:

/testcase/KEY`) that allows the AI to fetch a "Gold Standard" test case to use as a template. The AI inspects the live structure—including custom fields and folder paths—and ensures all new test cases created during the session match that exact schema. Example: A QA Lead asks the AI: "Look at PROJ-T123 as a template and create three new test cases for the updated Shopping Cart API." The AI fetches the template, copies the custom "Impact" and "Automation Status" fields, and applies them to the three new entries.

Use Case 3: Rapid Sprint Test Run Orchestration Problem: At the start of a sprint or release, QA managers must manually search for relevant test cases and bundle them into a "Test Run." This is time-consuming when dealing with hundreds of test cases spread across different folders. Solution: The AI can use search_test_cases_by_folder to identify all relevant tests for a feature and then execute create_test_run followed by add_test_cases_to_run to set up the testing environment in seconds. Example: A user tells the AI: "Create a new Test Run called 'Sprint 42 Regression' and add every test case currently located in the 'Core

Checkout' folder." The AI searches the folder, retrieves the keys, and generates the Test Run automatically.

Use Case 4: Synchronizing BDD Scenarios with Code Changes Problem: When developers change application logic (e.g., adding a new step to a multi-factor authentication flow), the BDD Gherkin scripts…

Key facts

  • Developer Tools & Code Intelligence, Project & Task Management
  • atlassian, jira, qa, testing, zephyr scale

Part of MCP Servers

Related MCP servers

  • MCP LaTeX Server — Create, edit, and manage LaTeX files. Requires an external LaTeX distribution like MiKTeX, TeX Live, or MacTeX.
  • MCP JSON — A collection of servers for file system operations, Google search, web automation, and executing terminal commands.
  • MCP Jupyter Complete — A server for Jupyter notebook manipulation with position-based operations and VS Code integration.
  • MCP LSP Go — An MCP server that connects AI assistants to Go's Language Server Protocol (LSP) for advanced code analysis.
  • MCP Kanban Memory — Manage complex AI agent workflows with a Kanban-based task management system.
  • MCP Manager — A full-stack application for managing Model Context Protocol (MCP) servers for Claude Desktop with a modern web interface.
  • AI Tools
  • Categories
  • Industries
  • CLI Coding Agents
  • MCP Servers
  • MCP Categories