Cypress Cloud MCP. Audit E2E test runs and quality metrics.
Gemini Works with every AI agent you already use
…and any MCP-compatible client
Just plug in your AI agents and start using Vinkius.
Cypress Cloud MCP Server audits your end-to-end (E2E) test suite. It lets your AI client monitor test runs, inspect specific failures, find flaky tests, and generate full performance reports.
You get full visibility into your app's quality metrics without leaving your chat window.
What your AI agents can do
Get instance
Gets full details for one spec file execution, including errors, screenshots, and video URLs.
Get instances
Lists all spec file executions within a run, returning IDs, names, and statuses.
Get run
Gets full details for a single Cypress Cloud run, showing passed/failed/pending counts and commit info.
Retrieves all organizational projects and their unique 6-character IDs on Cypress Cloud.
Lists recent test runs for a project, providing run IDs, commit info, branch, status, and duration.
Pulls full details for a single Cypress Cloud run, including total tests, passed/failed/pending counts, and commit info.
Retrieves error messages, screenshots, and video links for a specific spec file execution (instance).
Generates a report listing tests that fail intermittently, along with their flake rates and last failure dates.
Retrieves a report identifying the slowest tests by calculating average, p95, and maximum durations.
Fetches aggregated test result data across multiple runs, formatted for BI dashboards and audits.
Ask AI about this MCP
Supported MCP Clients
Gemini Waiting for input…
Cypress Cloud MCP Server: 10 Tools for E2E Testing
Use these tools to list projects, check run statuses, inspect failures, and generate performance reports from your Cypress Cloud environment.
019d7580get instance
Gets full details for one spec file execution, including errors, screenshots, and video URLs.
019d7580get instances
Lists all spec file executions within a run, returning IDs, names, and statuses.
019d7580get run
Gets full details for a single Cypress Cloud run, showing passed/failed/pending counts and commit info.
019d7580get runs
Lists recent test runs for a project, returning run IDs, statuses, and durations.
019d7580get tests
Lists individual tests inside a spec instance, providing titles, states, and error messages.
019d7580list projects
Lists all available projects and returns the unique 6-character project IDs.
019d7580report flaky
Generates a report that identifies tests that fail intermittently, including flake rates and last failure dates.
019d7580report runs
Gets an enterprise summary report of test runs for a given date range, suitable for BI dashboards.
019d7580report slow
Generates a report identifying the slowest tests by calculating average, p95, and maximum durations.
019d7580report tests
Gets a comprehensive report of individual test results across runs, including statuses and error messages.
Choose How to Get Started
Build a custom MCP for your own tools, or connect a ready-made integration from our catalog.
Build Your Own
Turn any API into an MCP. Import a spec, define Agent Skills, or deploy with MCPFusion.
- Import from OpenAPI, Swagger, or YAML specs
- Create Agent Skills with progressive disclosure
- Deploy to edge with MCPFusion framework
- Built in DLP, auth, and compliance on every call
- Real time usage dashboard and cost metering
- Publish to catalog or keep private
Make Your AI Do More
Start with Cypress Cloud, then connect any of our 4,700+ other servers whenever your AI needs more. One click, no limits.
- Use this MCP plus 4,700+ others, all in one place
- Add new capabilities to your AI anytime you want
- Every connection is secured and compliant automatically
- Track usage and costs across all your servers
- Works with Claude, ChatGPT, Cursor, and more
- New servers added to the catalog every week
What you can do with this MCP connector
You're gonna audit your end-to-end (E2E) test suite right from your chat window. The Cypress Cloud MCP Server lets your AI client monitor test runs, check specific failures, find flaky tests, and pull full performance reports. You get full visibility into your app's quality metrics without having to jump between dashboards.
List Project IDs lets you grab all your organizational projects and their unique 6-character IDs. Get Project Run Summary lists all recent test runs for a project, giving you the run ID, commit info, branch, status, and duration. Get Detailed Run Status pulls the full details for a single Cypress Cloud run, including total tests, passed/failed/pending counts, and commit info.
You can then Inspect Specific Test Instances to grab error messages, screenshots, and video links for a specific spec file execution. To find flaky tests, you use Identify Flaky Tests; it generates a report listing tests that fail intermittently, along with their flake rates and last failure dates. Get Performance Metrics runs a report identifying the slowest tests by calculating average, p95, and maximum durations.
For big picture audits, Generate Enterprise Reports fetches aggregated test result data across multiple runs, which is perfect for BI dashboards and compliance audits. Finally, List Project IDs gets all your organizational projects and their unique 6-character IDs.
How Cypress Cloud MCP Works
- 1 Subscribe to the Cypress Cloud Enterprise API and input your API Key and Project ID.
- 2 Your AI client sends a natural language request (e.g., 'List the last 5 failed runs').
- 3 The server executes the necessary tool (e.g.,
get_runs) and returns the data payload to your AI client.
The bottom line is that your AI client handles the API calls, so you talk to it like you're talking to a coworker.
Who Is Cypress Cloud MCP For?
This is for the QA Engineer who needs to stop jumping between Cypress dashboards and Jira to find a failure root cause. It's for the DevOps team that needs to prove CI reliability to management without running manual scripts. It's for the developer who needs test logs immediately in the IDE, not via a web portal.
Uses get_runs to check overall test status, and get_instance to grab specific error messages and videos when a test fails.
Runs report_flaky and report_slow to analyze CI/CD pipeline health, optimizing build reliability metrics.
Calls get_tests or get_instances to pull error logs and specs directly into the development workspace for quick debugging.
Uses report_runs and list_projects to get aggregated data for quarterly quality reports and team accountability.
What Changes When You Connect
- See exactly why a test failed. When you call
get_instance, you pull error messages, screenshots, and video URLs for a specific failing spec file. No more guesswork on flaky tests. - Track pipeline health across months. Use
get_runsto list recent test runs and see the pass/fail status, commit info, and duration instantly. Great for daily stand-ups. - Pinpoint the root cause of instability. Running
report_flakygives you a dedicated report showing which tests fail intermittently, plus their flake rates and last failure dates. - Evaluate performance trends. The
report_slowtool gives you test names and performance metrics (avg, p95, max durations), letting you know exactly where the slowdown is happening. - Audit large datasets for compliance. Use
report_runsorreport_teststo pull aggregated, structured data ready for BI dashboards or audit documentation. - Start from scratch. Call
list_projectsfirst to get the unique project ID you need before querying any other data.
Real-World Use Cases
The QA Engineer needs to debug a single failure.
A test failed, but the failure was vague. The QA engineer asks the agent to check instance 'ins_123'. The agent runs get_instance, returning the full error stack, a screenshot, and the video URL immediately. The failure is identified and fixed.
The DevOps team needs to prove CI stability.
The team needs to prove that the build process is stable and reliable for management. They ask the agent to run report_flaky for the last quarter. The agent compiles the report, identifying the top three flaky tests and their failure rates, proving where the CI needs work.
The Manager needs a quarterly quality report.
Quarterly reporting requires aggregated data on all test outcomes. The manager asks the agent to run report_runs for the last month. The agent fetches the aggregated summary report, providing structured data for the BI dashboard.
The Developer needs historical performance checks.
A new feature is slowing down the build. The developer asks the agent to run report_slow. The agent returns a report detailing which tests have the highest p95 and average durations, pinpointing the exact code area causing the slowdown.
The Tradeoffs
Trying to list everything manually
The developer tries to manually copy and paste run IDs and project names into a spreadsheet, hoping to track all test statuses. This is slow, error-prone, and misses critical metadata like commit info or branch.
→
Use list_projects first to get the project ID. Then, use get_runs to list all recent runs for that project. This gives you a structured list of IDs, commit info, and statuses in one go.
Asking for 'all failures' without scope
The user asks the agent, 'Show me all failures.' The agent must then prompt the user for a date range, a project ID, and a specific test instance ID, slowing down the conversation.
→
Be specific. Use report_tests for a general view of test errors, or use get_runs combined with get_instances if you know the approximate time window and project ID.
Assuming the API knows the project ID
The user simply says, 'Get the last run status.' The agent fails because the underlying system requires a specific project ID which the user hasn't provided.
→
When It Fits, When It Doesn't
Use this if you need to move beyond simple web UI views and need machine-readable data for analysis, or if you need to automate status checks for multiple projects. You need structured metrics (e.g., p95 duration, failure rates) that are hard to extract manually.
Don't use this if you just want to see the status of one test run—the web UI is faster. Don't use it if you only need a single, simple piece of data point. Use get_runs for a quick overview, but use report_flaky or report_slow when you need to analyze trends over time. If your goal is debugging a single failure, get_instance is the most direct tool.
Independent Platform Disclaimer: Vinkius is an independent platform and is not affiliated with, endorsed by, sponsored by, verified by, or otherwise authorized by Cypress. All third-party trademarks, logos, and brand names are the property of their respective owners. Their use on this website is strictly for informational purposes to identify service compatibility and interoperability.
VINKIUS INFRASTRUCTURE
Cloud Hosted
Managed infra
V8 Isolated
Sandboxed per request
Zero-Trust Proxy
No stored credentials
DLP Enforced
Policy on every call
GDPR Compliant
EU data residency
Token Compression
~60% cost reduction
Works with Claude, ChatGPT, Cursor, and more
The Model Context Protocol standardizes how applications expose capabilities to LLMs. Instead of operating in isolation, your AI gains direct access to external platforms, live data, and real-world actions through secure, standardized connections.
This server provides 10 capabilities that interface natively with Claude, ChatGPT, Cursor, and any MCP client. No middleware. No custom integration required.
Available Capabilities
Checking test results usually means copy-pasting and clicking through five different dashboards.
Today, finding the true cause of a test failure is a multi-tab, multi-system headache. You jump to Cypress, then to the CI dashboard, then maybe Jira, just to correlate the commit ID with the test run ID. You're sifting through raw logs, manually checking timestamps, and copy-pasting failure messages into a ticket—a process that takes forever.
With the Cypress Cloud MCP Server, you ask your agent. You say, 'Show me why the login spec failed on the main branch.' The agent runs `get_instance`, pulls the error stack, the screenshot, and the video links, and gives it to you in plain text. You get the answer, not a dashboard to look at.
Cypress Cloud MCP Server: Track Runs and Flaky Tests
Previously, tracking CI stability meant running ad-hoc scripts or waiting for a manual report. You'd have to correlate the date range with the project ID, then run a separate script just to check for flaky tests. It was a manual, multi-step process just to get an overview.
Now, you ask the agent to run `report_flaky` and `report_runs` for the last month. The agent handles the date filtering, the aggregation, and the report generation. You get a single, structured output listing all flaky tests and aggregated run summaries. It’s immediate, and it’s actionable.
Common Questions About Cypress Cloud MCP
How do I find the unique project ID using the Cypress Cloud MCP Server? +
You run the list_projects tool. This tool lists every project in your organization and gives you the unique 6-character ID you need for all other commands.
What is the best way to check for slow tests using the Cypress Cloud MCP Server? +
Use the report_slow tool. It identifies the slowest tests and returns metrics for average, p95, and maximum durations, helping you pinpoint performance bottlenecks.
Can I check the status of a run using the Cypress Cloud MCP Server? +
Yes, use the get_runs tool. This lists recent test runs, giving you the run ID, commit info, branch, and overall status (passed/failed/running).
What does the `get_instance` tool do? +
The get_instance tool retrieves detailed information for one specific spec file execution, including the error message, screenshots, and video links.
How do I check for flaky tests using the Cypress Cloud MCP Server? +
You run the report_flaky tool. It compiles a report showing tests that fail intermittently, along with their flake rates and last failure dates.
How do I get the full details of a specific test execution using the `get_instance` tool? +
The get_instance tool returns everything about one spec file run. You get the spec name, status, error messages, and links to screenshots and videos if the test fails.
Can I use `get_runs` to see the status of multiple test runs? +
Yes, get_runs lists recent test runs for a project. You get run IDs, commit info, branch, and overall statuses (passed/failed/running) for multiple runs at once.
What kind of data does `report_runs` provide for BI dashboards? +
The report_runs tool delivers aggregated run summaries. You get data like total tests, passed/failed/pending counts, and duration, formatted for use in business intelligence dashboards.
Can my agent help me debug a failed Cypress test? +
Yes. Use the 'get_instance' tool with the instance ID. The agent will retrieve explicit error messages and can provide you with direct links to failure screenshots and video recordings stored in Cypress Cloud.
How do I identify which tests are flaky in my project? +
Provide a start date to the 'report_flaky' tool. Your agent will analyze the historical data from Cypress Cloud and identify tests that intermittently pass and fail, helping you improve your CI reliability.
Can I use the agent to monitor the duration of my test runs? +
Absolutely. The 'report_slow' tool retrieves high-precision duration metrics. The agent can surface the slowest tests by average or p95 duration, allowing you to optimize your test suite's execution time.
Use it with your favorite AI tools
Connect this server to Cursor, Claude, VS Code, and more.
More in this category
Retool
Connect your AI assistant to Retool to inspect applications, audit users, review connected resources, and monitor workflows directly from chat.
DigitalOcean
Deploy and manage cloud infrastructure with simple virtual servers, managed databases, and Kubernetes clusters built for developers.
Forgejo (Gitea Fork)
Manage Forgejo and Gitea instances — check version compatibility, generate API tokens, and trigger CI/CD workflows directly from your AI agent.
You might also like
Candid (GuideStar)
Access comprehensive non-profit data via Candid — search organizations, verify charity status, and retrieve detailed financials directly from any AI agent.
Hyperswitch
Manage payments, customers, and refunds via Hyperswitch API.
Zendesk
Manage support tickets, users, and organizations on Zendesk — the complete customer service platform for high-scale teams.