humanize-with-rephrasy

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

• Potentially malicious external URL detected (medium risk: 0.30). The skill instructs the agent to install and execute an external MCP package (@rephrasy/mcp) from npm and make runtime API calls to rephrasy.ai infrastructure. As a third-party service/package lacking established official provenance or integrity guarantees, this constitutes a medium-tier unofficial operational dependency.