The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted user input via the /pw:generate command to create executable Playwright test scripts. This creates a potential vulnerability where malicious instructions in the user's specification or a targeted URL could be incorporated into the generated code.
Ingestion points: User stories, technical specifications, or external URLs provided as arguments to the /pw:generate command defined in SKILL.md.
Boundary markers: The instructions do not define explicit delimiters or instructions for the agent to ignore potentially malicious content embedded in the ingested data.
Capability inventory: The skill leverages the agent's ability to create and modify files (e.g., tests/auth/login.spec.ts) and execute shell commands (e.g., npx playwright test) as part of its documented workflow in SKILL.md.
Sanitization: The skill mitigates risks by providing a /pw:pw-review command that reviews generated tests for anti-patterns and quality, serving as a secondary validation step before code is finalized.

playwright-pro