video1.0<iframe src="https://www.loom.com/embed/8c360f6cdf984d809696e507421ce3ed" frameborder="0" width="2294" height="1720" webkitallowfullscreen mozallowfullscreen allowfullscreen></iframe>17202294Loomhttps://www.loom.com17202294https://cdn.loom.com/sessions/thumbnails/8c360f6cdf984d809696e507421ce3ed-74306253bd80d2c1-full.jpg469.538This Loom demonstrates how QA Wolf AI investigates failed tests and fixes underlying issues. The presenter creates a PR that introduces two app bugs and two outdated locators, updates the suite to add two locator failures, and then monitors the preview environment as QA.Wolf runs a suite with 37 workflows where 32 pass and 5 are running. After selecting PR34, they show that four flows fail across three attempts and that QA.Wolf begins active investigation, grouping failures by cause into sub-agents. The lead agent creates sub-agents, each with its own browser and repository, to identify root causes and make fixes.