The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill uses npx -y playwriter to ensure the Playwriter CLI is available, which fetches the package from the npm registry if not already installed.
[COMMAND_EXECUTION]: The skill executes the playwriter command-line utility to manage sessions and perform actions on the browser tab.
[REMOTE_CODE_EXECUTION]: The tool provides an evaluation flag (-e) that enables the agent to execute arbitrary JavaScript code within the context of the user's active browser tab, allowing for deep inspection and manipulation of page state.
[PROMPT_INJECTION]: The skill has an indirect prompt injection surface as it retrieves and processes untrusted content from the web pages the user visits. 1. Ingestion points: Data enters the agent's context through calls like page.url(), page.title(), and innerText via the playwriter -e command. 2. Boundary markers: The provided examples do not use delimiters or instructions to ignore embedded commands in the retrieved page data. 3. Capability inventory: The skill can execute local CLI commands and evaluate scripts in the browser. 4. Sanitization: There is no evidence of sanitization or filtering of the content retrieved from the browser before it is processed by the agent.

playwriter