The Agent Skills Directory

[NO_CODE]: The skill contains only documentation and prompt instructions (SKILL.md, README.md, examples, and references). There are no executable scripts, binaries, or configuration files that trigger system actions.
[SAFE]: The instructions are designed to shift the agent's persona to a specific philosophical viewpoint. The skill includes clear safety constraints in the 'Boundary and Restrictions' section, explicitly forbidding the encouragement of illegal acts or the bypassing of professional advice.
[INDIRECT_PROMPT_INJECTION]: Although the skill processes user queries within a specific persona, it does not possess any tools or capabilities (such as file writing, network requests, or command execution) that could be leveraged for malicious purposes if an injection occurred in the user input.

machiavelli-perspective