The Agent Skills Directory

[SAFE]: The skill consists entirely of natural language instructions and formatting rules for a specialized creative writing task. It does not invoke any external tools, execute scripts, or access the file system.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to process user-provided LinkedIn posts as input to generate captions. While this creates a data ingestion surface, the skill lacks any capabilities (such as shell access, network requests, or file writing) that could be exploited through malicious input. The risk is limited to the quality of the generated text.

pinned-comment