slidev-theme-neversink

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

• Third-party content exposure detected (high risk: 0.70). This skill includes an explicit data-ingestion workflow: assets/examples/generate_slides.py's load_data() fetches arbitrary HTTP URLs via requests.get and then the script parses and directly uses that untrusted API/JSON/CSV content (see generate_api_slides and generate_presentation) to build slide content, so external/user-provided webpages or API responses can influence the agent's output and behavior.