The Agent Skills Directory

[SAFE]: The skill consists entirely of Markdown instructions and internal reference documents for architectural guidance. No executable code, shell scripts, or binary files are present within the skill structure.
[SAFE]: The skill instructions explicitly incorporate security-conscious guardrails, directing the agent to refuse any design changes that would weaken trust-boundary validation, security controls, or data-loss prevention.
[SAFE]: No evidence of data exfiltration, hardcoded credentials, or unauthorized network operations was found. The skill operates within a conversational context without external dependencies.
[SAFE]: Analysis of the instructions for prompt injection or obfuscation revealed no malicious patterns. The use of emojis and standard Markdown formatting is consistent with the stated purpose of a design consultant tool.

duck-design