canvas-design
Pass
Audited by Gen Agent Trust Hub on Mar 30, 2026
Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The skill employs a fictional context technique by instructing the agent that 'The user ALREADY said "It isn't perfect enough..."'. This pattern is used to manipulate the agent's state into a refinement loop, overriding the actual conversation history.
- [EXTERNAL_DOWNLOADS]: The instructions permit the agent to 'Download and use whatever fonts are needed', which involves fetching remote assets from external sources to complete the design task.
- [COMMAND_EXECUTION]: The skill operates by having the agent generate and execute code to produce visual artifacts in
.pdfand.pngformats. It explicitly mentions refining the generated code to improve the final artistic output.
Audit Metadata