The Agent Skills Directory

[REMOTE_CODE_EXECUTION]: The skill fetches data from a remote endpoint at https://admin.hlx.page/config/${ORG}/sites.json and extracts values that are subsequently interpolated into shell commands. This pattern allows for remote command injection if the API response contains shell metacharacters (such as backticks or subshell syntax), as the values are expanded within a double-quoted string in a Bash execution context.
[COMMAND_EXECUTION]: Multiple shell commands are constructed using variables derived from external sources and user input. This includes the use of npx playwright open and node -e for dynamic script execution without input sanitization.
[EXTERNAL_DOWNLOADS]: The skill installs the playwright package and its Chromium browser dependency from the npm registry to facilitate browser-based login functionality.
[DATA_EXFILTRATION]: The skill captures and stores sensitive Adobe ID authentication tokens in the .claude-plugin/ directory. While this is the intended core functionality of the skill, storing raw credentials on the local filesystem represents a security risk. The skill attempts to mitigate exposure by adding the directory to the project's .gitignore file.

auth