Run visual E2E QA workflows with TestDriverAI
Use TestDriverAI to initialize, author, and run vision-assisted Vitest E2E checks for browser, desktop, extension, chatbot, OAuth, upload, iframe, canvas, and other hard-to-test flows.
npx skills add agentskillexchange/skills --skill run-visual-e2e-qa-workflows-with-testdriverai
Use TestDriverAI when an agent or QA operator needs to turn difficult user-interface flows into repeatable end-to-end tests that can run in cloud sandboxes with video replay. Upstream evidence frames it as a computer-use SDK for E2E QA testing, with a Vitest plugin, examples, an MCP server binary, and initialization that configures supported AI assistants such as VS Code, Cursor, and Claude Desktop.
What this skill actually does
The repeatable workflow is concrete: create a TestDriverAI account, run npx testdriverai init, let the initializer install Vitest and TestDriver, configure the API key and MCP assistant integration, write tests that use natural-language visual selectors and assertions, then run them with vitest run against spawned cloud devices. Invoke it instead of using a normal browser automation stack when flows depend on visual interpretation, desktop surfaces, third-party apps, extensions, OAuth, file uploads, iframes, canvas, video, or chatbot behavior. The scope boundary is agent-assisted QA test authoring and execution with Vitest and TestDriver, not a generic browser automation product card, cloud device platform, SDK listing, or hosted monitoring service.