The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes several commands to set up the environment and run tests:\n- Recommends installation of @playwright/cli via npm and playwright via pip.\n- Instructs the agent to start a local HTTP server using python -m http.server for testing local projects.\n- Executes local Python scripts (scripts/discover.py, scripts/run_tests.py, scripts/generate_report.py) to automate the QA workflow.\n- [EXTERNAL_DOWNLOADS]: The skill downloads software and browser binaries from a trusted provider:\n- Fetches the Playwright CLI and browser engines from Microsoft's official repositories.\n- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection (Category 8):\n- Ingestion points: Website content crawled during the discovery phase in scripts/discover.py.\n- Boundary markers: No explicit markers are used to delimit external content when generating test plans.\n- Capability inventory: The skill performs browser interactions, captures screenshots, and generates HTML reports via scripts/run_tests.py.\n- Sanitization: Web content is processed without filtering, allowing potentially malicious instructions on a website to influence the agent's subsequent testing actions.

test-web-ui