The Agent Skills Directory

[SAFE]: The skill is entirely instructional and restricted to generating structured markdown and Mermaid diagram syntax. No malicious patterns such as prompt injection, data exfiltration, or unauthorized command execution were detected.
[DATA_INGESTION]: The skill processes user-provided design briefs and challenges. While this represents a standard surface for indirect prompt injection, the skill lacks any dangerous capabilities (network access, shell execution, or sensitive file manipulation) that could be exploited by malicious input.

concept-sketching