GBIF MCP Server

A Model Context Protocol (MCP) server providing programmatic access to the Global Biodiversity Information Facility (GBIF) API.

Overview

This MCP server enables AI assistants and applications to interact with GBIF's extensive biodiversity data through the Model Context Protocol. GBIF provides access to hundreds of millions of species occurrence records, taxonomic information, and biodiversity research data from around the world.

Related MCP server: Aurora-MCP

What is GBIF?

The Global Biodiversity Information Facility (GBIF) is an international network and data infrastructure that provides open access to biodiversity data. The GBIF API offers programmatic access to:

• Species Data: Taxonomic information, species discovery, and name-matching utilities

• Occurrence Records: Indexed records of species observations and specimens

• Datasets: Information about published datasets and their sources

• Organizations: Data about publishing organizations and institutions

• Maps: Visualization of biodiversity data

• Literature: Peer-reviewed papers citing GBIF datasets

• Vocabularies: Standardized terminology for biodiversity data

What is MCP?

The Model Context Protocol is an open protocol that standardizes how AI applications interact with external data sources and tools. This server implements MCP to make GBIF's biodiversity data accessible to AI assistants like Claude.

Quick Start

# 1. Clone and build
git clone https://github.com/tyson-swetnam/gbif-mcp.git
cd gbif-mcp
npm install
npm run build

# 2. Add to Claude Code
claude mcp add gbif node "$(pwd)/build/index.js"

# 3. Test it
claude chat "Search GBIF for Panthera leo occurrences in Kenya"

Or for Claude Desktop, add to claude_desktop_config.json:

{
  "mcpServers": {
    "gbif": {
      "command": "node",
      "args": ["/ABSOLUTE/PATH/TO/gbif-mcp/build/index.js"]
    }
  }
}

Features

This MCP server provides access to key GBIF API endpoints with comprehensive parameter documentation:

• Species API: Search for species, get taxonomic information, and match scientific names

• Occurrence API: Query species occurrence records with 40+ advanced filters

• Registry API: Access information about datasets and publishing organizations

• Maps API: Generate visualizations of biodiversity data

• Literature API: Search for research papers citing GBIF data

• Validator API: Perform data quality checks

Comprehensive Parameter Descriptions

Every tool parameter includes:

• Detailed explanations of purpose and usage

• Real-world examples with actual GBIF data (e.g., "Example: 212 for family Felidae")

• Valid value lists for enums with human-readable descriptions

• Range constraints and data type information

• Links to GBIF API documentation for additional reference

This makes it easy for AI assistants to construct accurate queries without guessing parameter formats.

Response Size Limits

To ensure compatibility with AI context windows, this server implements automatic response size limiting:

• Maximum Response Size: 250KB (configurable via RESPONSE_MAX_SIZE_KB)

• Warning Threshold: 200KB (logs warnings for optimization)

• Smart Truncation: Large responses are automatically truncated with helpful pagination guidance

Configuration

Control response limiting via environment variables:

RESPONSE_MAX_SIZE_KB=250              # Maximum response size
RESPONSE_WARN_SIZE_KB=200             # Warning threshold
RESPONSE_ENABLE_TRUNCATION=true       # Enable smart truncation
RESPONSE_ENABLE_SIZE_LOGGING=true     # Log size metrics

Truncated Responses

When a response exceeds the size limit, you'll receive:

{
  "truncated": true,
  "originalSize": "1.2MB",
  "returnedSize": "248KB",
  "metadata": {
    "totalCount": 1000,
    "returnedCount": 23
  },
  "pagination": {
    "suggestion": "Use limit=20 with offset=0, then offset=20, offset=40...",
    "example": { "taxonKey": 212, "limit": 20, "offset": 0 }
  },
  "data": { ... }
}

Installation

From Source

# Clone the repository
git clone https://github.com/tyson-swetnam/gbif-mcp.git
cd gbif-mcp

# Install dependencies
npm install

# Build the server
npm run build

From NPM (when published)

npm install -g gbif-mcp

Using Docker

# Clone the repository
git clone https://github.com/tyson-swetnam/gbif-mcp.git
cd gbif-mcp

# Build the Docker image
docker build -t gbif-mcp:latest .

# Or use Docker Compose
docker-compose build

The Docker image uses a multi-stage build with Alpine Linux for a small footprint (~150MB).

Configuration

Option 1: Claude Desktop App

Add the server to your Claude Desktop configuration file:

macOS: ~/Library/Application Support/Claude/claude_desktop_config.json Windows: %APPDATA%\Claude\claude_desktop_config.json

{
  "mcpServers": {
    "gbif": {
      "command": "node",
      "args": ["/ABSOLUTE/PATH/TO/gbif-mcp/build/index.js"],
      "env": {
        "GBIF_USERNAME": "your-username",
        "GBIF_PASSWORD": "your-password"
      }
    }
  }
}

Important: Replace /ABSOLUTE/PATH/TO with the full path to your cloned repository.

Example (macOS):

{
  "mcpServers": {
    "gbif": {
      "command": "node",
      "args": ["/Users/yourname/projects/gbif-mcp/build/index.js"]
    }
  }
}

Example (Windows):

{
  "mcpServers": {
    "gbif": {
      "command": "node",
      "args": ["C:\\Users\\YourName\\projects\\gbif-mcp\\build\\index.js"]
    }
  }
}

After updating the configuration:

1. Save the file

2. Restart Claude Desktop

3. Look for the 🔌 icon to verify the MCP server is connected

Option 2: Claude Code CLI

If you're using Claude Code, you can add the MCP server using the CLI:

# Navigate to your gbif-mcp directory
cd /path/to/gbif-mcp

# Add the MCP server
claude mcp add gbif node build/index.js

# Or with environment variables for authenticated endpoints
claude mcp add gbif node build/index.js --env GBIF_USERNAME=your-username --env GBIF_PASSWORD=your-password

Verify the installation:

# List all MCP servers
claude mcp list

# Test the connection
claude mcp test gbif

Option 3: Docker Container

Run the MCP server in a Docker container for isolated, reproducible deployments:

Quick Start with Docker

# Build the image
docker build -t gbif-mcp:latest .

# Run the container (stdio mode for MCP)
docker run -i gbif-mcp:latest

Claude Desktop with Docker

Update your Claude Desktop configuration to use the Docker container:

macOS/Linux:

{
  "mcpServers": {
    "gbif": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "-e", "GBIF_USERNAME=your-username",
        "-e", "GBIF_PASSWORD=your-password",
        "gbif-mcp:latest"
      ]
    }
  }
}

Windows:

{
  "mcpServers": {
    "gbif": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "-e", "GBIF_USERNAME=your-username",
        "-e", "GBIF_PASSWORD=your-password",
        "gbif-mcp:latest"
      ]
    }
  }
}

Claude Code CLI with Docker

# Add the Docker-based MCP server
claude mcp add gbif docker run -i --rm gbif-mcp:latest

# With environment variables
claude mcp add gbif docker run -i --rm \
  -e GBIF_USERNAME=your-username \
  -e GBIF_PASSWORD=your-password \
  gbif-mcp:latest

Docker Compose

For easier management, use Docker Compose:

# Start the service
docker-compose up -d gbif-mcp

# View logs
docker-compose logs -f gbif-mcp

# Stop the service
docker-compose down

Edit docker-compose.yml to configure environment variables:

environment:
  - GBIF_USERNAME=${GBIF_USERNAME}
  - GBIF_PASSWORD=${GBIF_PASSWORD}
  - LOG_LEVEL=debug  # Change log level

Then use in Claude Desktop config:

{
  "mcpServers": {
    "gbif": {
      "command": "docker",
      "args": ["compose", "run", "--rm", "gbif-mcp"]
    }
  }
}

Docker Environment Variables

Pass environment variables to the Docker container:

# Using -e flags
docker run -i --rm \
  -e GBIF_USERNAME=myusername \
  -e GBIF_PASSWORD=mypassword \
  -e LOG_LEVEL=debug \
  -e CACHE_ENABLED=true \
  gbif-mcp:latest

# Using an env file
echo "GBIF_USERNAME=myusername" > .env.docker
echo "GBIF_PASSWORD=mypassword" >> .env.docker
docker run -i --rm --env-file .env.docker gbif-mcp:latest

Docker Image Details

• Base Image: Node.js 20 Alpine Linux

• Size: ~150MB (multi-stage build)

• User: Runs as non-root user (uid 1001)

• Security: Read-only root filesystem, no new privileges

• Health Check: Built-in health monitoring

Option 4: Other MCP Clients

For other MCP-compatible clients (Codex, Gemini CLI, etc.), add the server to the client's configuration file following their MCP server setup documentation:

General MCP Configuration Format:

{
  "mcpServers": {
    "gbif": {
      "command": "node",
      "args": ["/absolute/path/to/gbif-mcp/build/index.js"],
      "env": {
        "GBIF_USERNAME": "optional-username",
        "GBIF_PASSWORD": "optional-password"
      }
    }
  }
}

Environment Variables

Create a .env file in the project root for local development (optional):

# GBIF API Credentials (optional - only needed for downloads and authenticated endpoints)
GBIF_USERNAME=your-gbif-username
GBIF_PASSWORD=your-gbif-password

# GBIF API Configuration (optional - defaults shown)
GBIF_BASE_URL=https://api.gbif.org/v1
GBIF_USER_AGENT=GBIF-MCP-Server/1.0.0
GBIF_TIMEOUT=30000

# Rate Limiting (optional - defaults shown)
RATE_LIMIT_MAX_REQUESTS=100
RATE_LIMIT_CONCURRENT=10

# Caching (optional - defaults shown)
CACHE_ENABLED=true
CACHE_MAX_SIZE=100
CACHE_TTL=3600000

# Logging (optional - defaults shown)
LOG_LEVEL=info
LOG_FORMAT=json

Note: Most GBIF API endpoints do not require authentication. Credentials are only needed for:

• Requesting occurrence downloads

• Accessing private datasets

• Publishing data (not currently implemented)

Verifying Installation

After configuration, verify the server is working:

In Claude Desktop:

1. Open Claude Desktop

2. Look for the 🔌 MCP icon in the interface

3. Try a query: "Search GBIF for Panthera leo occurrences in Kenya"

In Claude Code CLI:

# List available tools
claude mcp tools gbif

# Test a simple query
claude chat "Use the gbif MCP server to search for Panthera leo"

Manual Test (stdio):

# Build and test the server directly
npm run build
echo '{"jsonrpc":"2.0","id":1,"method":"tools/list"}' | node build/index.js

Available Tools

The server provides the following MCP tools for interacting with GBIF data:

Species Tools

• gbif_species_search - Search for species with advanced filtering

• gbif_species_get - Get detailed information about a specific species

• gbif_species_suggest - Get species name suggestions for autocomplete

• gbif_species_match - Fuzzy match a species name against GBIF taxonomy

• gbif_species_children - Get direct taxonomic children of a taxon

• gbif_species_parents - Get complete taxonomic classification path

• gbif_species_synonyms - Get synonyms and alternative names

• gbif_species_vernacular_names - Get common names in multiple languages

• gbif_species_descriptions - Get textual descriptions of species

• gbif_species_distributions - Get geographic distribution information

• gbif_species_media - Get images, sounds, and videos

Occurrence Tools

• gbif_occurrence_search - Search occurrence records with 40+ filter parameters

• gbif_occurrence_get - Get complete details for a single occurrence

• gbif_occurrence_count - Fast counting without retrieving full records

• gbif_occurrence_verbatim - Get original unprocessed occurrence data

• gbif_occurrence_download_request - Request large dataset downloads (requires auth)

• gbif_occurrence_download_status - Check download status

• gbif_occurrence_download_predicate_builder - Build download filters

All tools include comprehensive parameter descriptions with examples. Use claude mcp tools gbif to see the full list.

Usage Examples

Once configured, you can use the MCP server through your AI assistant:

Species Queries

• "Search for occurrence records of Panthera tigris in India"

• "Find taxonomic information for the species Quercus robur"

• "What are the common names for Apis mellifera in different languages?"

• "Show me the complete taxonomic classification for the African lion"

• "Find all synonyms for the scientific name Felis leo"

Occurrence Queries

• "How many bird observations were recorded in California in 2023?"

• "Find preserved specimens of orchids collected before 1900"

• "Show me occurrence records with photos from National Parks"

• "Search for endangered species observations in Brazil"

Data Management

• "List datasets published by the Natural History Museum"

• "Show me recent papers citing GBIF data about pollinators"

• "Validate this biodiversity dataset before publication"

Troubleshooting

Server Not Connecting

Claude Desktop:

1. Check the configuration file path is correct

2. Verify you're using absolute paths (not relative paths like ~/ or ./)

3. Ensure the build directory exists: ls /path/to/gbif-mcp/build/index.js

4. Check Claude Desktop logs:

   • macOS: ~/Library/Logs/Claude/

   • Windows: %APPDATA%\Claude\logs\

Claude Code CLI:

# Check server status
claude mcp status gbif

# View logs
claude mcp logs gbif

# Remove and re-add
claude mcp remove gbif
claude mcp add gbif node /absolute/path/to/gbif-mcp/build/index.js

Build Errors

# Clean and rebuild
rm -rf build node_modules
npm install
npm run build

# Check for TypeScript errors
npm run build -- --noEmit

Docker Issues

Image Build Fails:

# Clean Docker cache and rebuild
docker system prune -a
docker build --no-cache -t gbif-mcp:latest .

# Check Docker version (requires 20.10+)
docker --version

Container Won't Start:

# Check container logs
docker logs <container-id>

# Run with debug logging
docker run -i --rm -e LOG_LEVEL=debug gbif-mcp:latest

# Test container health
docker run --rm gbif-mcp:latest node -e "console.log('test')"

Permission Issues:

# Ensure Docker daemon is running
docker ps

# Check if user has Docker permissions (Linux)
sudo usermod -aG docker $USER
# Log out and back in for group changes to take effect

Environment Variables Not Working:

# Verify environment variables are passed
docker run -i --rm gbif-mcp:latest node -e "console.log(process.env.GBIF_USERNAME)"

# Use --env-file for multiple variables
docker run -i --rm --env-file .env.docker gbif-mcp:latest

MCP Communication Issues with Docker:

• Ensure you're using -i (interactive) flag for stdio communication

• Use --rm to automatically remove containers after execution

• Don't use -d (detached) mode - MCP needs stdio connection

• Check that the command in your MCP config exactly matches: docker run -i --rm gbif-mcp:latest

Authentication Issues

If you're getting 401 errors for downloads:

1. Verify your GBIF credentials at GBIF.org

2. Check environment variables are set correctly

3. Create a .env file in the project root with your credentials

4. Note: Most endpoints don't require authentication - only downloads do

Rate Limiting

If you're being rate limited:

1. Reduce query frequency

2. Use the count endpoint before large searches

3. Consider caching results

4. Use occurrence downloads for large datasets instead of pagination

Node.js Version

Ensure you're using Node.js 18 or higher:

node --version  # Should be v18.0.0 or higher

API Coverage

This MCP server implements the following GBIF API sections:

• ✅ Species API

• ✅ Occurrence API

• ✅ Registry API

• ✅ Maps API

• ✅ Literature API

• ✅ Vocabularies API

• ✅ Validator API

Development

# Run in development mode
npm run dev

# Run tests
npm test

# Lint code
npm run lint

Resources

• GBIF API Documentation

• GBIF Website

• Model Context Protocol Documentation

• GBIF API Base URL

Contributing

Contributions are welcome! Please feel free to submit issues or pull requests.

1. Fork the repository

2. Create your feature branch (git checkout -b feature/amazing-feature)

3. Commit your changes (git commit -m 'Add some amazing feature')

4. Push to the branch (git push origin feature/amazing-feature)

5. Open a Pull Request

License

MIT License

Acknowledgments

This project uses the GBIF API to provide access to biodiversity data. GBIF is funded by governments and maintains a stable, version 1 API with backward compatibility guarantees.

Support

• GBIF API Issues: GBIF GitHub

• GBIF Community Forum: discourse.gbif.org

• MCP Issues: Open an issue in this repository