The Agent Skills Directory

[SAFE]: The skill provides documentation and code templates for browser automation with Playwright. All identified patterns across the 10 threat categories are benign or follow standard development practices.
[COMMAND_EXECUTION]: The skill documents standard Playwright CLI commands (e.g., npx playwright test, npx playwright show-trace) for local testing and debugging. These are legitimate tools for the intended testing workflow.
[EXTERNAL_DOWNLOADS]: The CI configuration example in references/configuration.md references official GitHub Actions (actions/checkout, actions/setup-node) and the standard Playwright installation command (npx playwright install). These originate from well-known, trusted sources and are standard for setting up test environments.
[CREDENTIALS_UNSAFE]: Code examples in references/page-object-model.md and references/configuration.md use placeholder credentials (e.g., user@test.com, password123) for demonstration purposes. These are clearly marked as example data and do not represent a credential leak.
[DATA_EXPOSURE]: The skill correctly demonstrates how to manage authentication state using Playwright's storageState feature (auth.json). This is a standard and documented feature for session persistence in automated testing.

playwright-expert