The Agent Skills Directory

[SAFE]: The skill consists entirely of natural language instructions and reference tables for text formatting and punctuation correction. There are no executable scripts, shell commands, or network operations included in the files.
[SAFE]: The instructions explicitly define protection rules to prevent the AI from altering technical content, including code fences (```), inline code (`), URLs (http/https), email addresses, and file paths.
[INDIRECT_PROMPT_INJECTION]: Although the skill is designed to process user-supplied text (an ingestion surface), it lacks any dangerous capabilities—such as file writing, shell access, or external network calls—that would allow an indirect prompt injection to escalate into a security risk. The potential impact is limited to incorrect text formatting in the output.

punct-master