The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill facilitates the generation and execution of test code based on user-provided requirements, which introduces a potential surface for processing malicious instructions.
Ingestion points: User-provided test scenarios and application requirements described during the interaction (SKILL.md).
Boundary markers: Absent. The instructions do not define delimiters or specific warnings to ignore instructions embedded within the test requirements.
Capability inventory: The skill utilizes CLI commands through Playwright and npm for testing and development (references/configuration.md).
Sanitization: Absent. There is no mechanism described for validating or escaping user-provided data before it is incorporated into generated scripts.
[METADATA_POISONING]: The 'author' field in the skill metadata refers to 'https://github.com/Jeffallan', which is inconsistent with the 'zroman88' author identity provided in the skill context.
[REMOTE_CODE_EXECUTION]: The skill recommends installing Playwright browsers and dependencies using 'npx playwright install --with-deps'. This is a standard procedure for the Playwright framework, which is a well-known service.

playwright-expert