Image Reader MCP

Overview

Image Reader MCP is a Model Context Protocol server that processes images with an evidence-first approach, combining Agent Media Twin capabilities with metadata extraction and citeable OCR evidence. Any MCP-compatible AI assistant can call its tools directly, letting the model decide when to use image analysis in conversations or agent runs.

It runs locally via the npm package @sylphx/image-reader-mcp, communicating over stdio on your machine. No authentication is required—install and run immediately. The current release is v0.1.0, published under the MIT license with source code available at SylphxAI/image-reader-mcp on GitHub.

Image Reader MCP is categorized under Development and integrates with MCP clients including Claude Desktop, Claude Code, Cursor, VS Code, Windsurf, Zed, and others. Maintainers at SylphxAI can claim this listing to verify ownership and keep information up to date.

Installation

// claude_desktop_config.json
{
  "mcpServers": {
    "image_reader_mcp": {
      "command": "npx",
      "args": [
        "-y",
        "@sylphx/image-reader-mcp"
      ]
    }
  }
}