d3-viz

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.80). SKILL.md's Data Handling explicitly permits fetching arbitrary URLs ("URL | Fetch with d3.csv(url) or d3.json(url)") and the Playground Integration and Prompt Output sections require the agent to analyze those fetched data (detect columns/types, populate controls, and generate natural-language prompts), so untrusted third‑party content would be read and could materially influence tool behavior and generated prompts.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.80). The skill's HTML/template loads and executes D3 at runtime from the external CDN URL https://cdn.jsdelivr.net/npm/d3@7 (and the ES module URL https://cdn.jsdelivr.net/npm/d3@7/+esm), which is a required runtime dependency that fetches and runs remote code in the user's browser.