The Agent Skills Directory

[SAFE]: No security issues detected. The skill consists entirely of markdown instructions and reference materials.
[NO_CODE]: This skill does not include any scripts, executables, or code files. It provides natural language instructions and decision-making logic for the agent to follow in its persona as a design auditor.
[PROMPT_INJECTION]: The skill processes user-supplied inputs such as mood words, audience snapshots, and brand hex codes. While these are entry points for untrusted data, the skill uses this information solely for generative creative tasks (designing palettes) and lacks dangerous capabilities like system command execution or network access, making the risk of indirect prompt injection negligible.

color-theory