
Vitest Midscene E2e
This brings AI-powered E2E testing to Vitest across web, Android, and iOS. Instead of writing brittle selectors, you describe what you want in natural language and let Midscene's aiAct API handle the planning and execution. The skill scaffolds new projects, converts existing ones, and helps you write, debug, and run tests. It's opinionated about structure (clones a boilerplate to ~/.midscene as the reference) and has one core rule: don't decompose user intent into granular API calls, just pass the whole instruction to aiAct. Works with Playwright for web, ADB for Android, and WebDriverAgent for iOS. Useful if you want less selector maintenance and more readable test code, though you're trading explicit control for AI interpretation.
- skills.sh installs
- 2.1K
- GitHub stars
- 321
- last updated
- Aug 21, 2026
Install
$ npx -y skills add web-infra-dev/midscene-skills --skill vitest-midscene-e2e --agent claude-codeInstalls into .claude/skills of the current project.
Files
Vitest Midscene E2E
Modules
| Module | Role |
|---|---|
| Vitest | TypeScript test framework. Provides describe/it/expect/hooks for test organization, assertions, and lifecycle. |
| Midscene | AI-driven UI automation. Interacts with UI elements via natural language — no fragile selectors. Core API: aiAct. |
Supported platforms:
- Web —
WebTest(Playwright Chromium):ctx.agent+ctx.page - Android —
AndroidTest(ADB + scrcpy):ctx.agentonly - iOS —
IOSTest(WebDriverAgent):ctx.agentonly
Workflow
Step 1: Clone boilerplate & ensure project ready
bash scripts/clone-boilerplate.sh
Sponsors
Details
- Repository
- github.com/web-infra-dev/midscene-skills
- Categories
- Testing & QA, DevOps & CI/CD, AI & Agent Building, Debugging, Mobile Development
- First seen
- May 16, 2026