Cypress Automatic Retries: Run Mode Limit Before Flaky Classification

Goal

I want to enable Automatic Retries in Cypress run mode so intermittent CI failures are retried, while tests that pass after a retry are flagged as flaky for review.

Constraints

Cypress retries re-execute the entire test function, including before and beforeEach hooks, and the retry count can be set separately for run and open modes. I need to know whether a global run-mode retry limit can surface flaky tests through the Cypress Dashboard or CLI without causing disproportionate pipeline execution time. Version-specific behavior should be confirmed against the installed release, and practical verification in a staging CI job is required before enabling globally.

Questions

• What is the maximum practical run-mode retry count before CI execution time grows disproportionately?
• Does a test that passes on a retry generate a dashboard alert or only a flaky label in run results?
• Can run-mode retries be scoped to specific suites without applying the same limit to interactive open mode?