You'd reach for this when you need to programmatically analyze images for signs of AI manipulation or deepfake artifacts. It runs forensic detection on faces and visual elements, looking for the kinds of anomalies that indicate synthetic or altered content. The server is hosted remotely on Vercel and connects via SSE transport, so you don't need to run detection models locally. Useful if you're building content moderation pipelines, verification workflows, or any system that needs to flag potentially manipulated imagery before it gets published or shared.
claude mcp add --transport sse io.github.evozim-deepfake-lens https://deepfake-lens-mcp.vercel.app/api/mcp