tvmaze

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

• Third-party content exposure detected (low risk: 0.10). The skill searches Tvmaze (tvmaze.com), an entertainment database where user-submitted or community-edited content may appear in search results, presenting a low indirect prompt injection risk via active search.

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

• Potentially malicious external URL detected (high risk: 0.70). The skill configures a remote Model Context Protocol (MCP) server via HTTP at https://unbrowse.ai/mcp/tvmaze.com, fetching instructions and tool definitions at runtime from an unverified third-party platform (Unbrowse), which constitutes an unverifiable runtime external dependency under a weak/unknown provenance tier.