The Agent Skills Directory

[SAFE]: No malicious patterns, credential theft, or unauthorized data access were detected. The skill is purely instructional, guiding the agent on how to structure text into a renderable Mermaid syntax diagram.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to process external 'raw material' such as user documents or notes. While this represents a surface for indirect prompt injection, the skill lacks any executable code, tool access, or network capabilities that could be exploited if malicious instructions were encountered in the input data. Boundary markers are not explicitly defined, but the risk is negligible due to the restricted output format.

mind-map