How do I use NexusForest MCP?

1. Click on "Install Server". 2. Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state. 3. In the chat, type @ followed by the MCP server name and your instructions, e.g., "@NexusForest MCP What was Brazil's forest loss in 2023?" That's it! The server will respond to your query, and you can continue using it as needed. Here is a step-by-step guide with screenshots.

NexusForest MCP

by danielberhane

Overview Schema Related Servers Score Discussions

Python

Local

NexusForest MCP

Bridge the gap between AI models and climate data - 24 years of Global Forest Watch insights covering 165+ countries, accessible through natural conversation.

NexusForest Model Context Protocol (MCP) server provides structured access to Global Forest Watch (GFW) deforestation and carbon emission data, enabling AI models to deliver quantified, source-attributed climate intelligence. Developed in partnership with Erasmus.AI to enhance ClimateGPT's capabilities with structured data access.

Features

Comprehensive Coverage: 165+ countries, 24 years (2001-2024)
Optimized Queries: Indexed SQLite database for efficient data retrieval
12 Specialized Tools: From simple lookups to complex trend analysis
Smart Architecture: Three-table design eliminates 60% data sparsity
Security-First: Parameterized SQL, input validation, read-only access
AI Integration: ClimateGPT API for enhanced climate insights

Related MCP server: Jana MCP Server

Quick Start

Prerequisites

Docker Desktop 4.0+
Claude Desktop (latest)
Excel data file: global_05212025.xlsx (Download from Global Forest Watch)

Installation

# 1. Clone repository
git clone https://github.com/danielberhane/nexusforest-mcp.git
cd nexusforest-mcp

# 2. Add Excel file to data/raw/
cp /path/to/global_05212025.xlsx data/raw/

# 3. Process data and start server
./rebuild-data.sh
./start-system.sh

Configure MCP Client

For Claude Desktop (macOS):

cat > ~/Library/Application\ Support/Claude/claude_desktop_config.json << 'EOF'
{
  "mcpServers": {
    "nexusforest": {
      "command": "/Applications/Docker.app/Contents/Resources/bin/docker",
      "args": ["exec", "-i", "nexus-mcp-server", "python", "-m", "nexus.mcp.mcp_stdio_server"]
    }
  }
}
EOF

For other MCP clients: See Installation Guide below

Windows: Use %APPDATA%\Claude\claude_desktop_config.json Linux: Use ~/.config/Claude/claude_desktop_config.json

Verify

Restart your MCP client and ask:

"What tools do you have available?"

You should see 12 nexusforest tools listed.

Example Queries

"What was Brazil's forest loss in 2023?"
"Compare deforestation in Indonesia vs Malaysia over the last decade"
"Show me the top 10 countries by carbon emissions from forests"
"Calculate the trend in global primary forest loss"

Installation Guide

macOS/Linux

# Clone and setup
git clone https://github.com/danielberhane/nexusforest-mcp.git
cd nexusforest-mcp
cp /path/to/global_05212025.xlsx data/raw/

# Build and run
chmod +x *.sh
./rebuild-data.sh  # Builds Docker images, processes data (~2 min)
./start-system.sh  # Starts MCP server

# Configure Claude (see Quick Start above)

# Daily usage
./start-system.sh  # Morning
./stop-system.sh   # Evening

Windows

Using PowerShell:

# Clone
git clone https://github.com/danielberhane/nexusforest-mcp.git
cd nexusforest-mcp

# Add data file
copy C:\path\to\global_05212025.xlsx data\raw\

# Build and run
docker-compose build
docker-compose --profile pipeline up
docker-compose --profile production up -d

# Configure Claude
notepad %APPDATA%\Claude\claude_desktop_config.json
# Add configuration (see docs/INSTALLATION.md)

Available Tools

Tool	Purpose	Example Query
`query_tree_cover_loss`	Get forest loss for specific country/year	"Brazil's 2023 forest loss"
`query_primary_forest`	Query virgin forest loss	"Primary forest in Congo"
`query_carbon_data`	Get CO2 emissions	"Carbon from Indonesia forests"
`analyze_trend`	Multi-year trends	"Brazil trend 2010-2023"
`compare_countries`	Country comparisons	"Brazil vs Indonesia"
`rank_countries`	Rankings by metric	"Top 10 by forest loss"
`calculate_primary_share`	Primary forest %	"Primary forest share Brazil"
`calculate_carbon_intensity`	CO2 per hectare	"Carbon intensity Amazon"
`compare_thresholds`	Compare across thresholds	"Compare 30% vs 75% Brazil"
`aggregate_global`	Global totals	"Global forest loss 2023"
`list_tropical_countries`	List tropical countries	"Show tropical countries"
`get_database_summary`	Database statistics	"Database summary"

See API Documentation for complete tool reference.

Architecture

MCP Client (Claude/ClimateGPT) → MCP Server (Docker) → SQLite Database (43MB)
                                      ↓
                                ClimateGPT API (enhanced analysis)

Three-table fact design optimized for query patterns
Docker containerization for consistent deployment
Metadata-driven validation for data quality
Index-optimized SQLite for sub-50ms queries

See Architecture Documentation for detailed system diagram and component design.

Data Updates

When new Global Forest Watch data arrives (annually):

# Replace Excel file
cp new_data_2026.xlsx data/raw/global_05212025.xlsx

# Rebuild database
./rebuild-data.sh

# Restart server
./start-system.sh

Project Structure

nexusforest-mcp/
├── src/nexus/
│   ├── mcp/                 # MCP server implementation
│   ├── data/                # ETL pipeline & database
│   └── config/              # Configuration
├── docker/                  # Container definitions
├── tests/                   # Test suite (49 tests)
├── data/
│   ├── raw/                # Input Excel files
│   └── processed/          # SQLite database (volume)
└── *.sh                    # Management scripts

Testing

# Run test suite
docker-compose --profile test up

# Expected output
✅ 49 tests passed

Security

✅ Parameterized SQL queries (injection-proof)
✅ Read-only database access
✅ Environment-based secrets
✅ Input validation & sanitization

License

Apache License 2.0 - See LICENSE for details.

Data Attribution

Source: Hansen/UMD/Google/USGS/NASA via Global Forest Watch

Author

Daniel Berhane Araya

Support

Source: Hansen/UMD/Google/USGS/NASA

This server cannot be installed

license - permissive license

quality - not tested

maintenance

How are these scores calculated?

Maintenance

–Maintainers

–Response time

–Release cycle

–Releases (12mo)

Commit activity

Resources

GitHub Repository

Need Help?

Related Servers

Unclaimed servers have limited discoverability.

Looking for Admin?

If you are the server author, to access and configure the admin panel.

Latest Blog Posts

Your AI Chatbot Just Exposed Your CEO's Salary to an Intern
By Om-Shree-0709 on July 2, 2026.
Agent Identity
MCP Security
OAuth Delegation
Why MCP Servers Need Execution Sandboxing (And Why Your Current Stack Isn't Enough)
By Om-Shree-0709 on June 30, 2026.
Agentic Ai
Prompt Injection
WebAssembly
Lightport: Open-Sourcing Glama's AI Gateway
By punkpeye on April 27, 2026.
OpenAI
open source

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/danielberhane/nexusforest-mcp'

If you have feedback or need assistance with the MCP directory API, please join our Discord server