canvas-design

Pass

Audited by Gen Agent Trust Hub on Sep 17, 2026

Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [PROMPT_INJECTION]: The skill uses a "simulated feedback" technique in the FINAL STEP section, instructing the model to act as if the user has already requested a refinement pass for a "masterpiece." This instruction is a form of prompt framing that overrides the actual conversation state to enforce specific quality standards.
  • [EXTERNAL_DOWNLOADS]: The skill provides license information for several open-source fonts and instructs the agent to download font files as needed if they are not available in the local directory. These references target well-known and trusted open-source projects, including repositories from Google Fonts, Vercel, and IBM.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 17, 2026, 12:28 AM