The Agent Skills Directory

[COMMAND_EXECUTION]: The run.js file utilizes execSync to perform system-level operations, specifically npm install and npx playwright install chromium during its automatic setup routine.
[REMOTE_CODE_EXECUTION]: The skill implements a dynamic execution pattern in run.js where arbitrary code received from user requests or external sources is wrapped in a template, written to a temporary file (.temp-execution-*.js), and then executed using the require() function.
[EXTERNAL_DOWNLOADS]: The skill is configured to download the playwright dependency from the NPM registry and fetch browser binaries from Microsoft's distribution servers during the setup phase defined in package.json and run.js.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it processes untrusted data from web pages using a high-privilege browser automation tool.
Ingestion points: User instructions for automation and the content of external websites accessed via Playwright (SKILL.md).
Boundary markers: Absent; there are no delimiters or specific instructions to the agent to treat external content as untrusted.
Capability inventory: The skill has full file system access (via fs), network access (via Playwright), and shell execution capabilities (via run.js and execSync).
Sanitization: No validation, sanitization, or sandboxing is applied to the generated scripts before they are executed.

playwright