Image MCP Server

Iris — Evidence-first image reading for AI agents — metadata, OCR text, regions, and citeable evidence without generative LLM.

Local serverstdio

What is the Image MCP server?

Image MCP server is a locally run integration for AI assistants that speak the Model Context Protocol. Iris — Evidence-first image reading for AI agents — metadata, OCR text, regions, and citeable evidence without generative LLM.

What you get

Evidence-first image reading for AI agents. One call turns any local image into an Agent Media Twin — dimensions, metadata, optional OCR with bounding boxes, and trust warnings you can cite without asking a vision LLM to guess.

What the assistant can call

Once Image is connected, these are the calls the assistant has available:

  • Local-first — read_image resolves paths on the local machine; no cloud vision API by default
  • Doc — Purpose
  • read_image — Read a local image and return dimensions, mime, metadata, optional OCR, and trust warnings
  • Channel — Status

Setting it up

Installation goes through your MCP client rather than a global install: point it at @sylphx/image-reader-mcp on npm and it is fetched when the client starts. The copy-paste blocks for Claude Desktop, Claude Code and Cursor are further down this page.

Choosing this one

This sits in the AI and media services group, where several servers overlap in what they claim to do but differ sharply once you actually set them up. Image's toolset — Local-first, Doc, read_image and 1 more — is a fair guide to whether it matches your workflow. It is maintained by shtse8; worth a glance at recent repository activity before you build anything load-bearing on it.

This entry was verified against Image's own documentation before publication; SyncDev keeps the directory reviewed rather than auto-generated.

Before you rely on it

  • It runs with your machine's permissions. That is convenient and also the reason to think about what you point it at before you approve a tool call.
  • MCP clients confirm each tool call by default. Leave that on until you have watched what the image mcp server does with a few real requests.

Available tools

ToolWhat it does
Local-firstread_image resolves paths on the local machine; no cloud vision API by default.
DocPurpose
read_imageRead a local image and return dimensions, mime, metadata, optional OCR, and trust warnings.
ChannelStatus

How to install the Image MCP server

{
  "mcpServers": {
    "image-reader-1": {
      "command": "npx",
      "args": ["-y", "@sylphx/image-reader-mcp"]
    }
  }
}

Add to claude_desktop_config.json, then restart Claude Desktop.

Example prompts to try

  • Use Image to Local-first.
  • Use Image to Doc.
  • Use Image to read image.

Frequently asked questions

It connects Image to MCP-compatible AI assistants such as Claude and Cursor, exposing 4 tools (Local-first, Doc, read_image, and more) that the assistant can call on your behalf. Instead of copying data back and forth by hand, the assistant works with Image directly.