We built CmdRunner after repeatedly seeing the same problem: software teams spend enormous time writing and maintaining test automation scripts, while manual test cases already exist in Excel sheets, documents, and test management tools. As AI started accelerating software development, testing became an even bigger bottleneck. Teams could build features faster than ever, but validating those changes still required significant manual effort or costly automation maintenance. We asked a simple question: What if teams could upload their existing test cases and have AI agents execute them directly in a browser like a human? That idea became CmdRunner. Instead of generating and maintaining Selenium, Playwright, or Cypress scripts, CmdRunner understands test intent, executes actions in a real browser, validates outcomes, and generates reports automatically. A major challenge with browser agents today is reliability. Many struggle with complex workflows, long-running tasks, and inconsistent execution. We spent significant time developing an execution approach that keeps agents grounded in the test objective, enabling them to navigate complex workflows more reliably while reducing the hallucinations and brittleness often seen in traditional AI-driven browser automation. We're excited to share CmdRunner with the Product Hunt community and would love your feedback.

CmdRunner

Upload test cases.AI agents test your web app automatically.

Upload test cases.AI agents test your web app automatically.