Wraps the Nature Vision API to identify species from images and return Latin scientific names with confidence scores. Exposes identification tools for eight categories including plants, birds, mammals, fungi, and insects. You'd reach for this when building natural history apps, educational tools, or anything that needs to recognize living organisms from photos and enrich the response with taxonomic data. Runs via stdio transport and requires a Nature Vision API key, though a rate-limited demo token is available. Configuration is straightforward through npx, making it simple to drop species identification capabilities into Claude or other LLM workflows without managing your own vision models.
# Nature Vision MCP Server
An **MCP (Model Context Protocol) server** that identifies biological species from images using the **Nature Vision API**, returning Latin names with confidence scores.
This server enables LLMs (such as Claude) to recognize species and enrich responses with biological knowledge.
## Features
- 🌿 Identify species from images (plants, animals, fungi, insects, etc.)
- 🧬 Returns **Latin scientific names** with confidence scores
- 🧠 Designed for **LLM tool usage** via Model Context Protocol (MCP)
- 🔌 Simple stdio-based MCP server
- 🚀 Easy to run with npx
- 🔐 API key via environment variables
## Supported Categories
- plant
- bug
- bird
- mammal
- reptile
- amphibian
- mollusc
- fungi
## Installation
npm install
**Configure in your MCP client:**
{
"mcp.servers": {
"nature-vision": {
"command": "npx",
"args": ["-y", "nature-vision-mcp"],
"env": {
"NATURE_VISION_API_KEY": "app-xxx"
}
}
}
}
For now, you can use token: app-dummy-token-2026, this will allow you request once per minute.
"NATURE_VISION_API_KEY": "app-dummy-token-2026"
npx nature-vision-mcp