This MCP server provides access to Cucumber Studio data through the Model Context Protocol. It exposes endpoints and tools that allow an AI assistant to list projects, retrieve project details, browse and search test scenarios by tags, access action words (reusable test steps), and inspect test executions and build information. The server supports both STDIO and HTTP transports, so you can connect via npx for quick local testing or run it inside a container for production deployments. Typical usage is to configure MCP clients to point at this server and then query tools like cucumberstudio_list_projects, cucumberstudio_get_project, cucumberstudio_list_scenarios, cucumberstudio_get_scenario, cucumberstudio_find_scenarios_by_tags, and cucumberstudio_get_project_runtimes or similar actions depending on the client integration.

Prerequisites:

• Node.js and npm installed on your machine
• Optional: Docker if you prefer containerized usage

Installation steps:

1. Clone the repository (or install via npm if published): git clone https://github.com/HeroSizy/cucumberstudio-mcp.git cd cucumberstudio-mcp

2. Install dependencies: npm install

3. Set up environment variables (create a .env file or use your shell environment): CUCUMBERSTUDIO_ACCESS_TOKEN=your_token CUCUMBERSTUDIO_CLIENT_ID=your_client_id CUCUMBERSTUDIO_UID=your_uid

   Optional overrides

   CUCUMBERSTUDIO_BASE_URL=https://studio.cucumberstudio.com/api MCP_TRANSPORT=stdio

4. Build the server (if you are developing or using TypeScript sources): npm run build

5. Run the server (example using npx as described in the README): npx cucumberstudio-mcp

6. Alternative: Docker usage

   • Pull and run the official image, supplying credentials via env vars docker run -e CUCUMBERSTUDIO_ACCESS_TOKEN=... -e CUCUMBERSTUDIO_CLIENT_ID=... -e CUCUMBERSTUDIO_UID=... herosizy/cucumberstudio-mcp

Tips and common considerations:

• Ensure your Cucumber Studio credentials are kept secret and not committed to source control.
• If using the HTTP transport, configure MCP_PORT and MCP_HOST as needed and ensure networking allows access.
• The server supports multiple transport modes (stdio, http, streamable-http). Choose based on your client integration.
• Enable logging options (LOG_LEVEL, LOG_API_RESPONSES, LOG_REQUEST_BODIES, LOG_RESPONSE_BODIES) to aid debugging, but disable verbose logs in production to improve performance.
• If using Docker, consider using a .env file and the env_file option in docker-compose for easier management.
• Verify required environment variables are present; missing credentials will cause API calls to fail.
• For desktop extension usage, the recommended flow is via the provided .dxt extension which encapsulates MCP configuration.