The Agent Skills Directory

[SAFE]: The skill provides legitimate guidance and code templates for Playwright. No obfuscation, persistence, or data exfiltration mechanisms were detected. Illustrative examples for authentication utilize generic test credentials that do not represent sensitive data leaks.
[COMMAND_EXECUTION]: Documents standard automation commands such as npx playwright test, npm run dev, and npx playwright show-trace within configuration and workflow templates.
[EXTERNAL_DOWNLOADS]: The CI/CD configuration templates reference official GitHub Actions (actions/checkout, actions/setup-node) and standard package management commands (npm ci, npx playwright install) to set up testing environments.
[PROMPT_INJECTION]: The skill documents capabilities for browser automation which interact with external web content, presenting a surface for indirect prompt injection.
Ingestion points: Interactions with external web pages and API responses via locators and route interception (documented in references/selectors-locators.md and references/api-mocking.md).
Boundary markers: Not explicitly included in the provided code templates.
Capability inventory: Generated code has extensive browser control, network interception, and local file system access for traces and reports (references/configuration.md).
Sanitization: The skill follows standard automation patterns without additional sanitization layers for web content processed by the agent.

playwright-expert