The Agent Skills Directory

[SAFE]: The skill consists of markdown documentation and LaTeX TikZ templates. Analysis confirms no executable scripts, network requests, or sensitive file access patterns are present in any of the files.
[SAFE]: The instruction to spawn a subagent for independent review is a standard agentic self-correction pattern and does not involve any instructions to bypass safety filters or override core behaviors.
[SAFE]: The LaTeX templates include example labels such as "System prompt" and "injection", but these are identified as static text literals within diagram nodes rather than functional code or prompt injection attempts.

tikz-flowchart