microsim-screen-capture

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

• Third-party content exposure detected (high risk: 0.70). The skill explicitly allows loading external CDN resources (e.g., vis-network.js, p5.js, Chart.js) referenced by the MicroSim's main.html and uses headless Chrome to render and produce screenshots that the agent may read/inspect, so it ingests public third-party content during its workflow.

MEDIUM W013: Attempt to modify system services in skill instructions.

• Attempt to modify system services in skill instructions detected (medium risk: 0.60). The skill explicitly instructs running Chrome with flags like --disable-web-security and --allow-file-access-from-files (bypassing browser security protections), which encourages disabling security mechanisms even though it doesn't request sudo or modify system files.