obsidian-canvas-creator

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

• Third-party content exposure detected (high risk: 0.80). The skill explicitly ingests and processes arbitrary user-provided content (e.g., "Turn this article into a canvas" + [article text]) and supports link nodes with external URLs, so the agent will read untrusted third-party/user-generated content at runtime.