canvas-design

Pass

Audited by Gen Agent Trust Hub on Sep 16, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to use user-provided input as the foundation for creating design philosophies and visual art, which can be exploited to influence agent behavior through embedded instructions.
  • Ingestion points: User input and 'subtle conceptual threads' are ingested in SKILL.md.
  • Boundary markers: No delimiters or instructions to ignore embedded commands are present.
  • Capability inventory: The skill creates and writes files (.md, .pdf, .png) and utilizes font rendering tools.
  • Sanitization: No sanitization or validation of the input is performed before processing.
  • [EXTERNAL_DOWNLOADS]: The skill instructions explicitly direct the agent to 'Download and use whatever fonts are needed,' which results in the agent accessing external network resources to fetch data files. This creates a risk if the agent retrieves resources from untrusted or malicious sources.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 16, 2026, 06:47 AM
Security Audit — agent-trust-hub — canvas-design