The Agent Skills Directory

[DATA_EXPOSURE]: The skill includes a function extract_strings that reads from source files to identify translatable content. This is a standard operation for localization workflows and does not involve accessing sensitive system files or credentials.
[PROMPT_INJECTION]: The llm_translate function interpolates user-provided text directly into an LLM prompt. While this creates an indirect prompt injection surface where a user could provide text containing instructions, the skill's capabilities are restricted to text processing and translation, posing minimal risk. The prompt uses labeled sections ('Source text:') to help delineate input.

Translation Assistant