The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructions require the user to disable the execution environment's security sandbox using the --sandbox danger-full-access flag, granting the agent full system permissions.
[COMMAND_EXECUTION]: Utilizes js_repl to run arbitrary JavaScript code on the host machine to manage Playwright sessions.
[COMMAND_EXECUTION]: Employs page.evaluate() and electronApp.evaluate() to dynamically inject and execute code within the renderer and main processes of the applications being tested.
[EXTERNAL_DOWNLOADS]: Downloads the playwright and electron packages from the NPM registry and fetches browser binaries via the npx playwright install command.

playwright-interactive