Advanced Visualization Techniques

Pass

Audited by Gen Agent Trust Hub on Apr 13, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill references several external libraries used for visualization, including Chart.js, Three.js, D3.js, Mermaid, and Tone.js.
  • The sources for these libraries are well-known and established Content Delivery Networks (CDNs) including cdnjs.cloudflare.com, esm.sh, cdn.jsdelivr.net, and unpkg.com.
  • These references are used for standard visualization functionality as intended by the skill's purpose.
  • [PROMPT_INJECTION]: The skill documents and encourages the use of a sendPrompt() global JavaScript function within the generated HTML widgets.
  • Ingestion points: Instructions in SKILL.md (Part 6) guide the agent to create buttons that execute sendPrompt('message') when clicked.
  • Boundary markers: The provided examples do not include delimiters or specific warnings to ignore embedded instructions within the prompts sent by this function.
  • Capability inventory: The sendPrompt() function allows the output of the skill to programmatically initiate a new interaction turn with the agent, effectively simulating user input.
  • Sanitization: There is no evidence of sanitization or validation logic to ensure that data interpolated into these prompts is safe or lacks malicious instructions. This mechanism represents a surface for indirect prompt injection where the generated UI can influence subsequent agent behavior.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 13, 2026, 08:53 PM
Security Audit — agent-trust-hub — Advanced Visualization Techniques