algorithmic-art
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill's HTML template fetches the
p5.jslibrary from Cloudflare's CDN and typography from Google Fonts. These are well-known, trusted services. - Evidence: In
templates/viewer.html, links tocdnjs.cloudflare.comandfonts.googleapis.comare used for script and style loading. - [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied conceptual foundations to generate generative art code and philosophies. This creates a surface where external input influences generated code, though it is limited to the artistic output.
- Ingestion points: User instructions provided as foundations in
SKILL.mdduring the philosophy creation phase. - Boundary markers: The instructions suggest using input as a foundation without constraining creative freedom, but do not mandate strict technical delimiters for user input.
- Capability inventory: The agent generates self-contained HTML and JavaScript artifacts in
templates/viewer.htmlintended for execution in an artifact viewer. - Sanitization: The skill instructions do not specify explicit sanitization or validation of the user-provided conceptual input.
Audit Metadata