Install Vision MCP Server

Pick your client, copy the command, done.

claude mcp add -e "OPENROUTER_API_KEY=${OPENROUTER_API_KEY}" vision-mcp -- vision-mcp

Required:OPENROUTER_API_KEY+ 1 optional

Environment Variables

Set these before running Vision MCP Server.

VariableDescriptionRequired

OPENROUTER_API_KEYYour API key from OpenRouterYes

OPENROUTER_MODELThe specific vision model to use from OpenRouterNo

Once configured, Vision MCP Server gives your AI agent access to:

analyze_imageAnalyzes images from URLs, file paths, or base64 data using OpenRouter vision models.

image_sourceprompt

After setup, try these prompts with your AI agent:

→Analyze the image at this URL and tell me what text is written on the sign.

→Look at this local file path and describe the layout of the UI elements.

→Can you identify the objects in this image and provide a JSON list of them?

→Analyze the provided image and explain the chart data shown.

Prerequisites & system requirements

An MCP-compatible client (Claude Code, Cursor, Windsurf, Claude Desktop, or Codex)
Node.js 18+ with npm/npx installed
OPENROUTER_API_KEY — Your API key from OpenRouter

Alternative installation methods

git clone https://github.com/TheNomadInOrbit/vision-mcp-server.git && cd vision-mcp-server && npm install && npm run build && npm install -g .

Save the docs, env vars, and workflow around Vision MCP Server in Conare so Claude Code, Codex, and Cursor remember it next time.