tiktok-launch-video

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

• Third-party content exposure detected (high risk: 0.90). The skill's required workflow explicitly instructs the agent to fetch and extract data from public site URLs (see SKILL.md Step 1c "WebFetch the page" and Step 1.5 "A site URL — we'll extract colors / fonts / voice"), and references/brand-identity.md similarly prescribes "WebFetch " and hyperframes capture <url> to read page HTML/CSS and screenshots, so untrusted third‑party page content will be parsed and can directly influence script, brand-kit generation, composition decisions, and subsequent tool actions.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill includes runtime fetching/execution of remote code — notably the GSAP script loaded from https://cdn.jsdelivr.net/npm/gsap@3.14.2/dist/gsap.min.js in the composition template, and the required npx hyperframes@latest commands which fetch and run package code from the npm registry — both are runtime external dependencies that execute remote code and are required for rendering.

MEDIUM W009: Direct money access capability detected (payment gateways, crypto, banking).

• Direct money access detected (high risk: 1.00). The skill explicitly includes TikTok Ads MCP tools that create campaign/ad_group/ad entities and accepts a "daily budget" when creating ad groups (tiktok_ads__tool_create_ad_group, tiktok_ads__tool_create_campaign, tiktok_ads__tool_create_ad). Managing/setting ad spend via the Ads API is a specific financial execution capability (ad budget updates/creation), so this meets the Direct Financial Execution criteria.