canvas-design

Pass

Audited by Gen Agent Trust Hub on Sep 14, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
  • [SAFE]: The skill is a set of creative instructions and font licenses. No malicious commands, obfuscation, or persistence mechanisms were detected.
  • [NO_CODE]: The skill does not include any executable scripts (Python, JavaScript, Shell, etc.). It functions purely through markdown-based instructions for the AI agent.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied text to generate design philosophies and visual artifacts, creating a standard vulnerability surface for potential instruction injection.
  • Ingestion points: User input processed during the 'Design Philosophy Creation' and 'Canvas Creation' steps in SKILL.md.
  • Boundary markers: No explicit delimiters or 'ignore embedded instructions' warnings are defined for user-supplied input.
  • Capability inventory: The skill utilizes file-writing capabilities to generate .md, .pdf, and .png outputs.
  • Sanitization: No specific sanitization or filtering logic is present for the user-influenced content.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 14, 2026, 07:19 AM
Security Audit — agent-trust-hub — canvas-design