The Agent Skills Directory

[DYNAMIC_EXECUTION]: The skill includes a dedicated executor (run.js) that dynamically generates and executes JavaScript code. It takes input from files, inline arguments, or standard input, wraps it in a template containing Playwright imports and helper utilities, writes it to a temporary file (.temp-execution-*.js) within the skill directory, and then executes it using the Node.js dynamic import() statement.\n- [COMMAND_EXECUTION]: The skill instructions in SKILL.md and the internal logic in run.js frequently utilize the Bash tool to perform environment setup, port scanning for local development servers, and to invoke the Node.js runtime for executing automation scripts.\n- [EXTERNAL_DOWNLOADS]: During initialization or the first run of the executor, the skill automatically downloads the playwright framework and required Chromium browser binaries. These downloads are performed using standard package managers (npm or bun) and target official registries and CDNs.\n- [INDIRECT_PROMPT_INJECTION]: As a browser automation tool, the skill is designed to ingest and process data from external web pages. Functions such as extractTexts and extractTableData in lib/helpers.js provide mechanisms for the agent to read content from arbitrary URLs. This creates a surface where instructions embedded in a target webpage could potentially influence the agent's behavior, as there are no explicit sanitization layers or boundary markers defined for the extracted content in the analyzed reference helpers.

playwright