html-ppt-zhangzara-capsule

Pass

Audited by Gen Agent Trust Hub on Jun 16, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill is a decorative and structural template for HTML-based presentations. It does not perform any sensitive file operations, network exfiltration, or unauthorized command execution. It includes a basic JavaScript implementation for slide navigation that operates entirely locally within the browser.
  • [EXTERNAL_DOWNLOADS]: The template references external assets from well-known services, specifically Google Fonts (fonts.googleapis.com and fonts.gstatic.com). These are standard resources for web development and do not execute remote code.
  • [PROMPT_INJECTION]: The skill processes user-supplied content (headlines, names, labels) to populate the slide deck. This creates a surface for potential indirect prompt injection if the user provides malicious content to be included in the template. This is the primary intended function of the skill.
  • Ingestion points: User-provided text substituted into placeholders in example.html.
  • Boundary markers: None present.
  • Capability inventory: Generates an HTML artifact with interactive navigation scripts.
  • Sanitization: The skill does not include specific sanitization logic; it relies on the model's output generation and user review of the resulting artifact.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 16, 2026, 11:02 AM
Security Audit — agent-trust-hub — html-ppt-zhangzara-capsule