Show HN: Qpilot – AI agent runs plain-text manual test cases in a real browser

Category: devtools

Tags: ai-agent, testing, playwright

Score: 6.3/10 (Innovation: 6, Technical: 6, Documentation: 7, Utility: 6)

Qpilot is an AI agent that executes plain-text manual test cases in a real browser, bridging the gap between human-readable test descriptions and automated execution. It uses AI models (Claude or OpenAI-compatible) and Playwright to interpret steps and handle UI changes via ARIA semantics, making it interesting for teams wanting to reduce manual testing without writing code.

Target audience: backend devs, frontend devs, qa engineers

Repository: https://github.com/broxhq/qpilot · TypeScript · MIT · 3 stars

View on Hacker News