skills/larchliu/md2x/canvas/Gen Agent Trust Hub

canvas

Pass

Audited by Gen Agent Trust Hub on Feb 16, 2026

Risk Level: LOW
Full Analysis
  • Prompt Injection (SAFE): No instructions found that attempt to override agent behavior or bypass safety filters.
  • Data Exposure (LOW): The skill defines 'file' and 'link' node types which can reference local file paths and external URLs. While part of the standard JSON Canvas specification, these represent a potential surface for exposing local paths in the agent's output.
  • Indirect Prompt Injection (LOW): The skill processes untrusted user data to generate visual diagrams. 1. Ingestion points: User text content mapped to node attributes. 2. Boundary markers: Output is constrained within a 'canvas' markdown code block. 3. Capability inventory: None (the skill has no execution or network capabilities). 4. Sanitization: No sanitization of user content is specified.
  • Remote Code Execution (SAFE): No remote scripts, package installations, or dynamic code execution patterns were identified.
Audit Metadata
Risk Level
LOW
Analyzed
Feb 16, 2026, 12:58 PM