tastefulkit

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

• Third-party content exposure detected (low risk: 0.10). The workflow reads design references and metadata from TastefulKit via MCP tools, which act as a searchable public registry and database of landing pages and components.

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

• Potentially malicious external URL detected (medium risk: 0.30). The skill connects to a hosted Model Context Protocol endpoint at https://tastefulkit.com/mcp/ to provide tool definitions and execution behavior, constituting a medium-tier common/reputable unofficial operational dependency.