# Cypress Cloud MCP > Cypress Cloud MCP connects your AI agent directly to your end-to-end testing pipeline. Ask questions about test runs, pinpoint specific failures, track intermittent flaky tests, and generate detailed performance reports without logging into a dashboard. It gives you full conversational control over your entire E2E quality lifecycle. ## Overview - **Category:** loved-by-devs - **Price:** Free - **Tags:** e2e-testing, test-automation, quality-assurance, debugging, test-reporting, ci-cd-pipeline ## Description Forget sifting through complex dashboards or exporting CSVs just to find out why a build failed. This MCP lets you talk to your testing results. You can ask your agent for an overview of recent runs, get the specific error messages and screenshots attached to a single failing test instance, or even request a report on flaky tests across your whole codebase. The system pulls detailed information—like slow test metrics or aggregated run summaries—and presents it back to you in plain language. Setting this up through Vinkius means you connect once, and your agent gets instant access to Cypress Cloud's full suite of audit tools. You get immediate visibility into performance gaps and quality trends, letting QA Engineers and DevOps teams stop guessing and start fixing the actual root cause. ## Tools ### list_projects Lists all projects on Cypress Cloud; this helps you find the unique ID needed to run other commands. ### get_runs Retrieves a summary of recent test runs, including their status and commit details. ### get_run Gets comprehensive data for a single run, detailing total tests, passed/failed counts, and duration. ### get_instances Lists every spec file execution within a given test run, showing statuses and durations. ### get_instance Gets deep details on one specific spec instance, including error messages, videos, and screenshots for failed tests. ### get_tests Lists individual tests inside a single spec file execution, providing titles, status, and errors. ### report_runs Generates an aggregated summary report of all test runs for a specific date range, useful for BI dashboards. ### report_tests Creates a detailed enterprise report listing individual tests and their statuses across the entire codebase. ### report_flaky Identifies intermittently failing tests, returning names, flake rates, and last known failure dates. ### report_slow Generates a report on the slowest tests in your suite, detailing average duration, p95 metrics, and maximum run times. ## Prompt Examples **Prompt:** ``` List the last 5 test runs for project 'abc123' ``` **Response:** ``` I found 5 recent runs for project 'abc123'. The latest run (ID: 101) finished with status 'Passed' on the 'main' branch. Run 99 failed with 2 failing specs. Would you like to see the details for the failed run? ``` **Prompt:** ``` Show me why instance 'ins_789' failed ``` **Response:** ``` Retrieving details for instance 'ins_789'... It failed during 'Login spec'. Error: 'Timed out retrying after 4000ms: Expected to find element: .login-success, but never found it.' I have the video and screenshot links available. ``` **Prompt:** ``` Give me a report of flaky tests starting from 2024-01-01 ``` **Response:** ``` Analyzing flaky tests since Jan 1st... I identified 3 flaky tests. 'Checkout flow' has a 15% flake rate, and 'User profile update' has flaked twice this month. I can list the last flake dates for these. ``` ## Capabilities ### List all organization projects Retrieves a list of organizational Cypress Cloud projects and their required unique IDs. ### Monitor test run statuses Provides status details for recent project runs, including pass/fail counts and commit information. ### Inspect failed test instances Pulls deep details on a specific failing spec file execution, providing error logs, video URLs, and screenshots. ### Identify intermittent failures Generates reports showing which tests fail randomly (flaky) and tracks the last time they failed across your code base. ### Audit performance metrics Retrieves dedicated reports detailing test duration averages, p95 scores, and slow tests for CI/CD review. ## Use Cases ### Debugging a critical failure immediately after merge. A developer sees a test fail in the CI log. Instead of manually getting the run ID and then navigating to the instance, they ask their agent to 'Show me why instance X failed.' The MCP uses get_instance to pull the error logs, screenshots, and video links directly into the chat. ### Reviewing quarterly stability for management. An Engineering Manager needs a high-level view of quality. They prompt the agent to 'Generate an enterprise report on test results since Q1.' The MCP runs report_tests, providing aggregated data ready for a presentation slide. ### Improving CI reliability by finding hidden issues. A DevOps Engineer suspects tests are failing randomly. They ask the agent to 'Run a flaky test audit starting from last month.' The MCP uses report_flaky, giving them names and historical flake dates to prioritize fixes. ### Investigating performance bottlenecks. A team notices build times are creeping up. They ask the agent for 'a report on slow tests last week.' The MCP uses report_slow, providing specific metrics (avg/p95) and identifying which exact test methods need optimization. ## Benefits - Debugging goes from clicking through dashboards to asking a question. Instead of manually cross-referencing run IDs and specs, you prompt your agent for the specific error messages or video links needed to diagnose a failure. - Stop treating flaky tests like random bugs. Use the report_flaky tool to automatically identify intermittent failures across the whole codebase and track exactly when they last occurred. - Get full performance visibility without exporting data. The report_slow tool pulls average, p95, and max durations instantly, giving you concrete metrics for optimizing your CI/CD pipeline reliability. - Audit compliance effortlessly. You can request a comprehensive run summary (report_runs) or individual test result data (report_tests), formatted right away for management review or BI dashboards. - Reduce context switching dramatically. Your agent lets you list projects and then drill down, moving from an overall status check (get_runs) to viewing granular error logs (get_instance) all in one chat thread. ## How It Works The bottom line is you skip all the clicks. You ask the question, and the MCP pulls the necessary data from Cypress Cloud. 1. Subscribe to the Cypress Cloud MCP on Vinkius. 2. Enter your enterprise API Key and Project ID into your agent client. 3. Prompt your AI client with a natural language question, like 'Show me why the checkout process failed last week.' ## Frequently Asked Questions **How do I find the right Project ID using Cypress Cloud MCP?** You start by calling list_projects. This tool retrieves a full list of all organizational projects, giving you the unique 6-character IDs needed to run any other report or query. **Can Cypress Cloud MCP tell me if a test is flaky?** Yes. You use the report_flaky tool. It analyzes your history and generates reports identifying tests that fail intermittently, along with their last known failure dates. **What tools does Cypress Cloud MCP give for performance checking?** It provides two main reporting tools: report_slow identifies the slowest methods by metrics like p95 duration, while get_run gives overall run summaries and counts that indicate general stability. **If a test fails, how do I use Cypress Cloud MCP to see the evidence?** First, you might call get_runs for context. Then, use get_instances with the specific instance ID; this tool pulls the error messages, screenshots, and video URLs right into your chat. **Is Cypress Cloud MCP only for recent runs?** No. You can generate historical reports using report_runs or report_tests by providing a specific start date (YYYY-MM-DD) to cover large time ranges for compliance audits.