The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands (open or xdg-open) on a file path derived from user-provided input. While it mandates sanitization of the key to prevent shell injection, the execution of shell commands on computed paths is a notable capability.
[EXTERNAL_DOWNLOADS]: The generated HTML output references CSS from Google Fonts (fonts.googleapis.com), a well-known and trusted service, to style the translation table interface.
[PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface (Category 8) because it ingests and processes untrusted English text from $ARGUMENTS to generate translations. Malicious instructions could be embedded in this text to attempt to override agent behavior during the translation process.
Ingestion points: $ARGUMENTS variable in SKILL.md.
Boundary markers: No delimiters or boundary markers are used to isolate user input.
Capability inventory: Writing files to the local .tmp directory and executing shell commands via open or xdg-open in the terminal environment.
Sanitization: The skill specifies sanitizing the filename for shell safety but does not include sanitization or filtering for the content of the translations themselves.

localise