gaia-test-e2e

AI-automated E2E testing orchestrates deterministic Playwright or Cypress test execution with LLM-based quality assessment. It combines tool outputs with AI judgment on stability and coverage to produce a final verdict, ensuring functional and non-functional requirements are met during deployment.

LLM-based quality assessment in E2E testing evaluates deterministic test results against predefined rubrics for stability, coverage, and root cause analysis. It uses a configured LLM to judge test outcomes, combining findings with tool outputs to resolve a final verdict like APPROVE or REQUEST_CHANGES.

Yes, you can use Cypress test suites with AI automated E2E testing through tool adapter contracts. The automation supports both Playwright and Cypress adapters to execute deterministic E2E tests, requiring access to compatible test suites and a configured LLM for judgment.

Verdict resolution in automated E2E testing combines deterministic tool output from Playwright or Cypress with LLM findings based on predefined rubrics. This combined evidence produces a final verdict of APPROVE, REQUEST_CHANGES, or BLOCKED to update review gates in deployment workflows.

Yes, a configured LLM is required for automated E2E test workflows to perform LLM-based judgment on test results. The LLM evaluates stability, coverage, and root cause analysis, combining its findings with deterministic tool outputs to generate a final verdict.