The Agent Skills Directory

[SAFE]: The skill does not contain any malicious patterns, obfuscation, or persistence mechanisms. It focuses on providing comprehensive guidance for legitimate browser automation tasks.
[COMMAND_EXECUTION]: A utility script (scripts/playwright_setup.py) uses subprocess.run() to manage the installation of Playwright browser binaries. This is a standard and expected operation for setting up the Playwright environment and does not involve shell execution of untrusted input.
[CREDENTIALS_UNSAFE]: The instructions explicitly promote secure credential management, advising users to utilize environment variables or .env files and providing clear warnings against hardcoding secrets in scripts.
[EXTERNAL_DOWNLOADS]: The skill references the official Playwright project and standard Python packages (playwright, pytest, pytest-asyncio) available through official registries. No downloads from untrusted or unknown remote sources were identified.
[PROMPT_INJECTION]: The instructions use standard natural language for guiding agent behavior without attempting to bypass safety filters or override system constraints.
[DATA_EXFILTRATION]: While the skill enables network interaction via a browser, it does so for its primary purpose of automation and testing. There is no evidence of patterns designed to exfiltrate sensitive local data to unauthorized external domains.

browser-automation