tiktok-collection-scraper

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

• Third-party content exposure detected (high risk: 1.00). The skill directly fetches and parses untrusted public social media content — it issues requests to TikTok pages and redirects (e.g., https://www.tiktok.com/@{username}, vm.tiktok.com short links) and to TikTok APIs (/api/user/collection_list/, /api/collection/item_list/) and then reads HTML/JSON (see resolve_user/_parse, get_collections, get_items) for arbitrary user-provided targets, exposing the agent to third-party user-generated content that could contain injected instructions.