The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill uses npx to dynamically download and run the @playwright/cli package from the official NPM registry at execution time, as defined in scripts/playwright_cli.sh.
[COMMAND_EXECUTION]: The skill includes a shell wrapper script (scripts/playwright_cli.sh) designed to execute Playwright CLI commands, which includes capabilities for browser interaction and executing arbitrary JavaScript within a browser context.
[PROMPT_INJECTION]: The skill's primary purpose is to interact with and extract data from external websites, creating a surface for indirect prompt injection where malicious instructions on a webpage could influence the agent's behavior.
Ingestion points: The skill frequently extracts page content using snapshot, eval, and textContent (e.g., in SKILL.md and references/workflows.md).
Boundary markers: There are no explicit instructions or delimiters provided to the agent to treat extracted web content as untrusted data.
Capability inventory: The agent has the ability to execute shell commands via the Playwright wrapper and can run arbitrary JavaScript in the browser context via eval or run-code.
Sanitization: No sanitization or filtering logic is specified for the data retrieved from external URLs before it is processed by the agent.

playwright