You'd reach for this when you need to analyze whether text was written by a human or generated by AI. It exposes a scoring system that evaluates writing style to detect AI-generated content, giving you a measure of how "original" versus "generated" a piece of text appears. The server runs remotely via SSE transport, so you can integrate it without local setup. Useful for content moderation workflows, academic integrity checks, or any scenario where you need to programmatically flag potentially AI-written material. The scoring approach means you get nuanced results rather than just a binary yes/no answer.
claude mcp add --transport sse io.github.evozim-ghostblock https://ghostblock-mcp.vercel.app/api/mcp