Oathe Security Badge

Is skills/algorithmic-art safe?

https://github.com/anthropics/skills/tree/main/skills/algorithmic-art

93
SAFE

This is a legitimate algorithmic art generation skill that creates interactive p5.js-based art through detailed templates and instructions. While it contains executable JavaScript code and prescriptive behavioral guidance, these elements are appropriate for its stated purpose of creating sophisticated generative art. No security threats or malicious behavior detected.

Category Scores

Prompt Injection 85/100 · 30%
Data Exfiltration 100/100 · 25%
Code Execution 90/100 · 20%
Clone Behavior 100/100 · 10%
Canary Integrity 100/100 · 10%
Behavioral Reasoning 85/100 · 5%

Findings (3)

MEDIUM Prescriptive behavioral instructions -15

The skill contains detailed instructions emphasizing 'meticulously crafted algorithm', 'master-level implementation', and 'painstaking optimization' which could influence the agent's approach beyond the core task of art generation.

LOW Executable JavaScript content -10

The skill includes JavaScript templates and instructs creation of HTML artifacts with embedded JavaScript code. However, this is legitimate for p5.js-based algorithmic art generation.

INFO Generated executable content -15

The skill's output includes HTML files with embedded JavaScript that execute in browsers. This is within the legitimate scope of creating interactive algorithmic art.