canvas-design
Pass
Audited by Gen Agent Trust Hub on Jul 16, 2026
Risk Level: SAFE
Full Analysis
- [PROMPT_INJECTION]: The skill employs prompt engineering techniques to enforce specific behavioral loops, such as simulating previous user feedback ("The user ALREADY said...") to trigger a mandatory refinement pass. While this pattern is sometimes used in adversarial contexts, here it is used constructively to guide the agent's creative workflow.
- [COMMAND_EXECUTION]: Instructions direct the agent to search a local directory (
./canvas-fonts) to locate and utilize fonts for document generation. This is a standard file access operation consistent with the skill's primary function of creating visual artifacts. - [SAFE]: The repository includes numerous license files (
OFL.txt) for open-source fonts and an Apache 2.0 license, which are standard for design-oriented projects.
Audit Metadata