The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute Playwright tests and associated CLI tools (e.g., vpx playwright test, vpx playwright show-report) to verify application behavior.
[DYNAMIC_EXECUTION]: Standard Playwright functions like page.addInitScript() and page.evaluate() are used in SKILL.md to interact with the browser's JavaScript environment, such as setting localStorage for test setup.
[INDIRECT_PROMPT_INJECTION]: The skill possesses an inherent attack surface as it is designed to ingest and process project-specific files.
Ingestion points: Project configuration (playwright.config.ts), test specifications (tests/playwright/**/*.test.ts), and external fixture data.
Boundary markers: Absent; the agent relies on standard context interpolation for processing project files.
Capability inventory: Shell command execution via vpx (SKILL.md), network operations through APIRequestContext (SKILL.md), and file system access for test generation.
Sanitization: Present; the skill strongly recommends and provides patterns for runtime schema validation of API responses using the valibot library.
[DATA_EXFILTRATION]: The skill includes a proactive security measure in evals/files/harness/global.fixtures.ts that implements a network guard using page.route(), effectively blocking any outgoing requests that do not match the defined application base URL origin.

playwright-e2e-testing