amazon-tiktok-signal

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

• Third-party content exposure detected (medium risk: 0.30). The workflow reads video search results and metadata returned from the external KOLSprite MCP tool (asin_video_search), which includes outsider-authored video titles, captions, and creator metadata.

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

• Potentially malicious external URL detected (medium risk: 0.30). The configuration declares an operational MCP server dependency via an external HTTPS URL (https://mcp.kolsprite.com/mcp) with third-party community provenance, which is fetched and used at runtime to execute skill functions.