Test smarter with Snagly: 30 open-source QA skills for AI coding agents
Snagly is a free, MIT-licensed collection of 30 skills for AI coding agents, designed to address the gap where AI agents can drive a browser but lack testing discipline. It works with GitHub Copilot, Claude Code, Cursor, Codex, and 70+ other agents. The author demonstrated Snagly by running it against their production site, softwaretestingtrends.com, without any prior fixes. The tool found eleven issues, including a critical accessibility bug on the signup page. Notably, one of its findings was incorrect, which the author highlighted as significant. The full walkthrough, from an empty folder to running against production, is available as a video. Snagly is described as turning 'an AI that can drive a browser' into 'an AI that tests like a QA professional.' A skill is defined as a reusable instruction set that teaches the agent a specific working method.
Snagly provides a free, open-source way to give AI coding agents structured QA skills, potentially improving test reliability.