The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: Fetches the Playwright framework, browser binaries, and project dependencies from official npm and Microsoft registries during initialization and setup.
[COMMAND_EXECUTION]: Utilizes shell commands to initialize project infrastructure, execute test suites, capture debugging traces, and run local validation hooks during the development process.
[DATA_EXFILTRATION]: Offers optional reporting features to push test summaries and results to well-known external platforms like Slack, TestRail, and BrowserStack using user-configured environment variables.
[PROMPT_INJECTION]: The skill processes user-provided scenarios, application code, and external URLs to generate Playwright tests, which establishes a surface for indirect prompt injection.\n
Ingestion points: User stories in the /pw:generate command, codebase exploration, and external URL analysis.\n
Boundary markers: No explicit markers are defined in the instructions to isolate user-provided input within the code generation templates.\n
Capability inventory: The skill possesses capabilities for file writing (via the Edit tool) and shell command execution (via the Bash tool).\n
Sanitization: There is no documented automated sanitization or filtering of user-provided input before its use in generating executable test scripts.

playwright-pro