canvas-design
Pass
Audited by Gen Agent Trust Hub on Apr 7, 2026
Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
- [PROMPT_INJECTION]: The skill uses simulated user history to steer the agent's behavior, explicitly stating 'The user ALREADY said "It isn't perfect enough..."'. This technique is used to force compliance with specific aesthetic standards by fabricating prior interactions.
- [PROMPT_INJECTION]: The instructions direct the agent to actively mask its nature by ensuring work looks 'human-crafted' and 'not AI-generated', which is a directive aimed at bypassing typical AI behavioral markers.
- [EXTERNAL_DOWNLOADS]: The skill instructs the agent to 'Download and use whatever fonts are needed' to complete the task, implying network operations to unspecified external sources.
- [COMMAND_EXECUTION]: The instructions require the agent to search a local directory ('./canvas-fonts') and utilize file-creation capabilities to output .md, .pdf, and .png files.
Audit Metadata