What does the server output?
It outputs two files: a TypeScript Page Object class (`.ts`) and a Cypress test suite (`.cy.ts`) containing positive, negative, workflow, accessibility, performance, and edge case tests.
@jprealini
Basic information
With an already setup Cypress framework, pass the URL of the page you need to create a Page Object class and test cases for, and it will create the page object class with all locators, action methods (type, click), get methods, and main workflow methods, and a test file (cy.ts or
Add this server to your MCP-compatible client using the configuration below.
{
"mcpServers": {
"cypress-mcp": {
"command": "npx",
"args": [
"-y",
"@jprealini/cypress-mcp"
]
}
}
}Crear archivos Page Object y Tests directamente en el proyecto Cypress
Generar Page Object y Tests de Cypress
This MCP server automatically generates complete Cypress Page Object classes and comprehensive test suites for any web page by scraping the page with Puppeteer and parsing HTML with Cheerio. It is for developers who want to quickly create structured Cypress test infrastructure.
Install via npm, then start the server with npx tsx main.ts. Use an MCP client to call the generateLocator tool with a URL parameter; it returns both a TypeScript Page Object class and a Cypress test suite.
It outputs two files: a TypeScript Page Object class (`.ts`) and a Cypress test suite (`.cy.ts`) containing positive, negative, workflow, accessibility, performance, and edge case tests.
The server requires Node.js, and its dependencies include @modelcontextprotocol/sdk, puppeteer, cheerio, zod, and typescript.
Yes, Puppeteer runs in headless mode and waits for network idle to capture dynamically loaded content.
The server uses the Model Context Protocol (MCP) with a tool-based interface; no specific authentication is mentioned.
The server only generates tests based on the current page state; it may not capture every dynamic interaction. Error handling includes invalid URLs and network failures.
OrangePro local-first CLI + MCP server for behavior mapping, grounded test generation, and dynamic proof.

AI SEO tools for MCP clients — audit whether AI crawlers can read a website, validate JSON-LD and robots.txt, and generate llms.txt, AI-crawler-aware robots.txt, and JSON-LD schema without leaving the conversation.

MCP server for AI Agents — exposes Yocoolab feedback threads and design selections as tools

SSH MCP server for AI agents: remote commands, file transfer, log search and server audits through OpenSSH.

Comments