Show HN: Qpilot – run your manual test cases without writing test code

Category: devtools

Tags: ai-agent, testing, browser-automation, playwright, qa

Score: 7.0/10 (Innovation: 7, Technical: 7, Documentation: 8, Utility: 6)

Qpilot is an AI agent that executes plain-text manual test cases directly in a real Chrome browser using Claude or any OpenAI-compatible model. It removes the need to write or maintain test code by reading the page's accessibility tree at each step, and it pauses for human input when it hits OTPs or captchas. The combination of LLM-driven browser control with pragmatic handling of real-world testing friction makes it a compelling tool for teams that rely on manual QA.

Target audience: QA engineers, frontend developers, product managers, and backend developers involved in web app testing

Repository: https://github.com/broxhq/qpilot · TypeScript · MIT · 13 stars

View on Hacker News