algorithmic-art

Pass

Audited by Gen Agent Trust Hub on May 14, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTIONREMOTE_CODE_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The templates/viewer.html file includes a reference to the p5.js library via cdnjs.cloudflare.com. This is a standard and well-recognized Content Delivery Network used for distributing legitimate software libraries.
  • [PROMPT_INJECTION]: The SKILL.md instructions utilize reinforced prompting techniques, such as the 'CRITICAL GUIDELINES' for craftsmanship and the 'DEDUCING THE CONCEPTUAL SEED' section. These directives instruct the agent to adopt a high-expertise persona and incorporate subtle artistic references into its output. While these use strong language common in prompt engineering, they are entirely aligned with the skill's stated artistic purpose and do not attempt to bypass security filters.
  • [REMOTE_CODE_EXECUTION]: The skill's primary function is to generate and execute JavaScript within an HTML environment to produce art. This is a functional requirement of the skill and is implemented through clear, parameterized templates that emphasize reproducibility and safety through seeded randomness.
Audit Metadata
Risk Level
SAFE
Analyzed
May 14, 2026, 12:14 AM
Security Audit — agent-trust-hub — algorithmic-art