Connects to OpenAI's vision models to analyze images from URLs and return detailed descriptions. Exposes a single describe-image tool that takes an image URL and hands back text describing what's in the photo. You'd reach for this when you need Claude to understand image content without building your own OpenAI vision integration. Requires an OpenAI API key set as an environment variable. Built in TypeScript, ships as an npm package, runs via stdio transport. Handles the usual error cases like invalid URLs and API failures. Straightforward if you're already paying for OpenAI and want image analysis in your MCP workflow without extra plumbing.
claude mcp add --transport stdio mcp-s-ai-image-recongnition-mcp uvx image-recongnition-mcp