Unsplash MCP Server

> A Model Context Protocol server that provides Unsplash photo search and retrieval capabilities. This server enables LLMs to search, retrieve, and get random photos from Unsplash's extensive collection. This is a Go implementation that offers tools like search_photos, get_photo, and random_photo.

✨ Features

Advanced Image Search: Search Unsplash's extensive photo library with filters for:
Keyword relevance
Color schemes
Orientation options
Custom sorting and pagination
Detailed Photo Information: Get comprehensive details about specific photos including EXIF data, location, and photographer information
Random Photo Selection: Get random photos with flexible filtering options
Multiple Connection Modes:
Standard I/O mode for direct integration with LLMs
Server-Sent Events (SSE) mode for web-based connections

Available Tools

search_photos - Search for photos on Unsplash
query (string, required): Search keyword
page (number, optional): Page number (1-based), default: 1
per_page (number, optional): Results per page (1-30), default: 10
order_by (string, optional): Sort method (relevant or latest), default: "relevant"
color (string, optional): Color filter (black_and_white, black, white, yellow, orange, red, purple, magenta, green, teal, blue)
orientation (string, optional): Orientation filter (landscape, portrait, squarish)

get_photo - Get detailed information about a specific photo
photoId (string, required): The photo ID to retrieve

random_photo - Get one or more random photos
count (number, optional): The number of photos to return (Default: 1; Max: 30)
collections (string, optional): Public collection ID('s) to filter selection. If multiple, comma-separated
topics (string, optional): Public topic ID('s) to filter selection. If multiple, comma-separated
username (string, optional): Limit selection to a specific user
query (string, optional): Limit selection to photos matching a search term
orientation (string, optional): Filter by photo orientation. Valid values: landscape, portrait, squarish
content_filter (string, optional): Limit results by content safety. Valid values: low, high
featured (boolean, optional): Limit selection to featured photos

Installation

Option 1: Download Pre-built Binary

Download the latest pre-built binary for your platform from the GitHub Releases page and follow the installation instructions below.

<details>
<summary><b>macOS Installation</b></summary>

macOS with Apple Silicon (M1/M2/M3):