The Agent Skills Directory

[DYNAMIC_EXECUTION]: The run.js file implements a universal executor that takes arbitrary JavaScript code from command-line arguments or standard input, wraps it in a template, writes it to a temporary file in the skill directory, and executes it using require(). This facilitates the runtime execution of dynamically generated scripts.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to interact with and extract data from external websites, which makes it susceptible to instructions embedded in third-party content.
Ingestion points: Content from web pages accessed via page.goto(), extractTexts(), and extractTableData() in lib/helpers.js.
Boundary markers: There are no boundary markers or specific instructions to ignore embedded commands when processing external site data.
Capability inventory: Full browser control, file system writes (for screenshots and execution scripts), local port discovery via detectDevServers in lib/helpers.js, and general network access.
Sanitization: The skill does not filter or sanitize retrieved web content before processing.
[COMMAND_EXECUTION]: The setup script and run.js perform environment configuration by executing shell commands such as npm install and npx playwright install.
[EXTERNAL_DOWNLOADS]: The skill downloads the Playwright library and browser binaries from public registries during its setup phase.

playwright-skill