Qpilot is a new open-source tool that uses AI to run manual test cases written in plain English step-by-step in a real browser, eliminating the need to write or maintain automation code. It supports popular AI models (Claude by default, or any OpenAI-compatible endpoint) and handles edge cases like captchas by pausing and asking the user, making it accessible to non-programmers and more resilient to UI changes than traditional script-based testing.
Summaries like this, in your inbox every morning.
Sign up free →What happened
Qpilot is a new tool that lets users paste plain-text test cases and an AI agent executes them step-by-step in a real Chrome browser, watching for passes, failures, or warnings. If the agent encounters a captcha or one-time password, it pauses and asks the user directly before continuing.
Why it matters
Manual testing often requires either hiring humans to click through UIs or writing and maintaining fragile automation scripts in languages like Selenium or Playwright. Qpilot removes both barriers—anyone can write a test by describing steps in English, and the agent reads the page like a human using accessibility semantics, so it survives UI layout changes that would break traditional scripts.
What to watch
The tool requires Node.js 20.12+, Google Chrome, and an API key from Anthropic (Claude, the default) or any OpenAI-compatible model endpoint (Qwen, Ollama, corporate gateways, etc.). Users can run test batches from folders of markdown files and view live results and timing; finished runs are stored under Recent runs on the home page.
No comments yet. Be the first to share your thoughts!
Log in to join the discussion
Get curated AI news from 200+ sources delivered daily to your inbox. Free to use.
Get Started FreeFree · takes 30 seconds · unsubscribe anytime
1 minute a day. The AI essentials.
200+ sources · Email / LINE / Slack
