The Agent Skills Directory

[SAFE]: The skill functions as a stylistic text regenerator and does not contain any malicious patterns or unauthorized operations. It strictly uses the provided knowledge base to inform its creative outputs.
[DATA_EXPOSURE]: The skill uses the Read tool to access brand-specific documentation within its knowledge/styles/ directory. The instructions include a validation step that ensures only the 13 supported style files can be loaded, effectively preventing path traversal or unauthorized file access.
[COMMAND_EXECUTION]: While the skill defines allowed-tools such as Glob and Grep, these are standard platform utilities and are not used in a manner that would allow for arbitrary command execution or system compromise based on the provided logic.
[PROMPT_INJECTION]: The skill processes user-supplied text from $ARGUMENTS. While it does not utilize explicit boundary markers (like XML tags) to isolate this input, the prompt contains clear phases for parsing, analysis, and validation that direct the model to treat the input as data for stylistic transformation rather than instructions to be executed.

brand-quill