The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes the subprocess.run module in scripts/build_app.py and scripts/install_pake.py to execute system version checks and the pake command-line tool. The implementation correctly passes arguments as a list rather than a single shell string, which is a security best practice that prevents shell injection vulnerabilities.
[EXTERNAL_DOWNLOADS]: During initialization, the skill identifies and installs the pake-cli package globally using standard package managers (npm or pnpm). This download is explicitly documented in the skill's instructions and is required for its primary purpose of converting web pages to applications.
[INDIRECT_PROMPT_INJECTION]: The skill accepts user-provided input such as URLs, application names, and advanced options (like file injection). While the skill's own execution logic is secure, it provides an attack surface where an agent could be instructed to package a malicious URL or include sensitive local files (via the --inject parameter) into the generated application. This risk is inherent to the tool's function as a developer utility.

web-to-app