The Agent Skills Directory

[SAFE]: The skill is primarily instructional and provides a methodology for creative writing and world-building. It does not utilize any dangerous tools, perform network requests, or execute arbitrary code. All references are to internal documentation files.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to audit user-provided or file-based power system descriptions, which creates a surface for indirect prompt injection. However, because the skill lacks capabilities to perform file writes, network exfiltration, or command execution, the potential impact of such an injection is negligible. The absence of specific boundary markers for external data is noted but does not escalate the risk in this context.

power-system