Security audit
Security checks for vulnerabilities and agentic risk
This skill is a disclosed generative-art helper with minor documentation and branding concerns, but no evidence of hidden access, persistence, exfiltration, or destructive behavior.
Before installing, understand that this skill creates browser-based generative art files and generated HTML may load p5.js and fonts from external CDNs. If you plan to publish or share outputs, review or remove the Anthropic-styled branding unless you have appropriate permission.
Early documentation states the skill outputs .md, .html, and .js files, implying separate artifacts. Later sections explicitly require embedding all JavaScript inline in one HTML file and say not to create separate .js files, which directly contradicts the earlier output description.
The skill mandates copying a branded templates/viewer.html verbatim and preserving Anthropic branding exactly, while presenting the result as original algorithmic art. This creates a provenance and attribution risk: generated outputs may misleadingly imply official Anthropic authorship, endorsement, or a right to reuse branded UI assets, especially if shared externally.
No suspicious patterns detected.
