d3-viz

Pass

Audited by Gen Agent Trust Hub on Aug 21, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill references the official D3.js library via its official CDN at d3js.org.\n- [PROMPT_INJECTION]: The templates provided in SKILL.md and assets/interactive-template.jsx use d3.selection.html() to render tooltips based on data properties, creating a surface for indirect prompt injection.\n
  • Ingestion points: Data properties such as d.label in SKILL.md and assets/interactive-template.jsx which are interpolated into tooltips.\n
  • Boundary markers: Absent; no explicit markers or instructions are provided to the agent to ignore potentially malicious content within the data.\n
  • Capability inventory: The skill is restricted to DOM and SVG visualization rendering; no file-write or network exfiltration capabilities are present in the provided scripts (verified in SKILL.md and assets/interactive-template.jsx).\n
  • Sanitization: Absent; the implementation uses .html() for data interpolation without sanitizing or escaping the content, allowing arbitrary HTML/script injection from the data source.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 21, 2026, 05:04 AM
Security Audit — agent-trust-hub — d3-viz