AllTrails MCP Server

⚠️ DEPRECATED - PROJECT ARCHIVED

This project has been deprecated and is no longer functional.

Date: January 25, 2026
Reason: This project violated AllTrails' Terms of Service by scraping their website without authorization.

At the request of AllTrails' CTO, all scraping functionality has been disabled. This repository is maintained for historical purposes only and to preserve GitHub stars, but the code no longer works and should not be used.

What Happened?

AllTrails contacted me regarding this repository, citing:

Unauthorized scraping/data extraction in violation of their Terms of Service
Potential intellectual property violations
Redistribution of copyrighted content without a license

Alternatives

If you need trail data, please:

Use AllTrails' official website
Contact AllTrails directly about API access
Respect website Terms of Service and copyright

Original Description (For Historical Reference)

A Model Context Protocol (MCP) server that ~~provides access to AllTrails data~~ [DEPRECATED - NO LONGER FUNCTIONAL].

Features

🥾 Search trails by national park
📍 Get detailed trail information including difficulty, length, elevation gain, and descriptions
🏔️ Comprehensive trail data from AllTrails including ratings, route types, and summaries
🤖 Seamless Client integration via MCP protocol

Technical Specifications

Protocol Support

Protocol Version: MCP 1.9.4
Communication: Standard input/output (stdio)
Capabilities: Tools
Server Name: alltrails-mcp
Server Version: 0.1.0

Tools Available

`search_trails`

Search for trails in a specific national park using AllTrails data.

Parameters:

park (required): Park slug in format us/state/park-name (e.g., us/tennessee/great-smoky-mountains-national-park)

`get_trail_details`

Get detailed information about a specific trail by its AllTrails slug.

Parameters:

slug (required): Trail slug from AllTrails URL (the part after /trail/)

Installation Options

Option 1: With Virtual Environment (Recommended)

This approach isolates dependencies and prevents conflicts with other Python projects.

1. Clone the Repository

Code

git clone <your-repo-url>
cd alltrails_mcp

2. Create Virtual Environment

Code

python3 -m venv alltrails_mcp_venv
source alltrails_mcp_venv/bin/activate  # On Windows: alltrails_mcp_venv\Scripts\activate

3. Install Dependencies

Terminal

pip install -r requirements.txt

4. Verify Installation

Test that the server starts without errors:

Code

python3 server.py

You should see the server start without crashing. Press Ctrl+C to stop.

Prerequisites

Python 3.8 or higher
This server can be used with any MCP-compatible client. I use Claude Desktop with Pro subscription.
macOS (tested) or other Unix-like system

Option 2: With System Python

If you prefer not to use a virtual environment, you can install dependencies globally.

1. Clone the Repository

Code

git clone <your-repo-url>
cd alltrails_mcp

2. Install Dependencies Globally

Terminal

pip install -r requirements.txt

3. Verify Installation

Test that the server starts without errors:

Code

python3 server.py

You should see the server start without crashing. Press Ctrl+C to stop.

Option 3: With Docker

This approach runs the server in an isolated container.

1. Build the Image

Terminal

docker build -t alltrails-mcp .

2. Run the Container

Terminal

docker run -i --rm alltrails-mcp

3. Use with Claude Desktop

For Docker-based deployment, you can configure Claude Desktop to use the containerized server:

config.json

{
  "mcpServers": {
    "alltrails_mcp_server": {
      "command": "docker",
      "args": ["run", "-i", "--rm", "alltrails-mcp"]
    }
  }
}

Claude Desktop Configuration

1. Locate Claude Desktop Config

Find your Claude Desktop configuration file:

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

If it doesn't exist, create a json file named claude_desktop_config.json in the above directory.

2. Add MCP Server Configuration

Add the following to your claude_desktop_config.json file:

Option A: Using Virtual Environment (Recommended)

config.json

{
  "mcpServers": {
    "alltrails_mcp_server": {
      "command": "/path/to/your/alltrails_mcp/alltrails_mcp_venv/bin/python3",
      "args": ["/path/to/your/alltrails_mcp/server.py"]
    }
  }
}

Option B: Using System Python

config.json

{
  "mcpServers": {
    "alltrails_mcp_server": {
      "command": "python3",
      "args": ["/path/to/your/alltrails_mcp/server.py"]
    }
  }
}

Or with absolute Python path:

config.json

{
  "mcpServers": {
    "alltrails_mcp_server": {
      "command": "/usr/bin/python3",
      "args": ["/path/to/your/alltrails_mcp/server.py"]
    }
  }
}

Important: Replace /path/to/your/alltrails_mcp with the actual absolute path to your project directory.

Other MCP Clients

For other MCP-compatible clients, refer to their documentation for server configuration. The server implements the standard MCP protocol and should work with any compliant client.

Server Command: python3 server.py
Communication: Standard input/output (stdio)
Protocol Version: MCP 1.9.4

3. Find Your Python Path

For Virtual Environment Users:

Code

cd /path/to/your/alltrails_mcp
source alltrails_mcp_venv/bin/activate
which python3

For System Python Users:

Code

which python3

Use the output path in your configuration.

4. Install Dependencies

If Using Virtual Environment:

Dependencies are already installed in your virtual environment from step 3.

If Using System Python:

Install dependencies globally:

Terminal

pip install -r requirements.txt

5. Restart Claude Desktop

Completely quit and restart Claude Desktop for the changes to take effect.

Usage Examples

Once configured, you can use these commands in Claude Desktop:

Search for Trails

By park name:

Code

Find trails in Great Smoky Mountains National Park

By location:

Code

What are the best hiking trails in Yosemite?

With specific criteria:

Code

Show me moderate difficulty trails in Yellowstone

Using park slugs directly:

Code

Search for trails in us/california/yosemite-national-park

For specific activities:

Code

Find family-friendly trails in Zion National Park

Get Trail Details

By trail name:

Code

Get details for Alum Cave Trail to Mount LeConte

Using trail slugs:

Code

Get details for trail us/tennessee/alum-cave-trail-to-mount-leconte

For planning purposes:

Code

I need detailed information about Rainbow Falls Trail including difficulty and elevation

Combination Queries

Search and get details:

Code

Find the most popular trails in Grand Canyon National Park and give me details about the top rated one

Compare trails:

Code

Search for trails in Great Smoky Mountains and tell me which ones are best for beginners

Trip planning:

Code

I'm visiting Yellowstone for 3 days. Find me a mix of easy and moderate trails with good views

Natural Language Examples

The MCP server works with natural language, so you can ask questions like:

"What are some good day hikes in the Smoky Mountains?"
"Find me a challenging trail with waterfalls in Tennessee"
"I want to hike to a summit with 360-degree views"
"Show me trails that are good for photography"
"Find dog-friendly trails in national parks"
"What's the difficulty level of Charlies Bunion trail?"

Common Park Slugs

Great Smoky Mountains: us/tennessee/great-smoky-mountains-national-park
Yosemite: us/california/yosemite-national-park
Yellowstone: us/wyoming/yellowstone-national-park
Grand Canyon: us/arizona/grand-canyon-national-park
Zion: us/utah/zion-national-park

Troubleshooting

Server Not Connecting

Check the logs:
bash
```
tail -f ~/Library/Logs/Claude/mcp.log
```

Verify your config file:

bash

cat ~/Library/Application\ Support/Claude/claude_desktop_config.json

Test the server manually:

bash

cd /path/to/your/alltrails_mcp
source alltrails_mcp_venv/bin/activate
python3 server.py

Common Issues

"Connection closed" errors: Usually indicates a Python path or virtual environment issue

Path configuration issues: Check that all paths in the config are absolute and correct

Import errors: Ensure all dependencies are installed in the correct Python environment (virtual environment vs system Python)
Python path errors: Use which python3 or which python to verify the correct Python executable path

Debug Mode

For detailed debugging, check the MCP logs:

Code

# macOS
tail -f ~/Library/Logs/Claude/mcp.log

# The server also outputs debug information to stderr

Development

Running Tests

Run the test suite using pytest:

bash

pytest

Linting

Format and check code using ruff:

bash

ruff check .
ruff format .

Project Structure

Code

alltrails_mcp/
├── app/
│   └── alltrails_scraper.py    # AllTrails scraping logic
├── examples/
│   └── claude_desktop_config.json  # Example configuration file for Claude Desktop
├── tests/                      # Test suite
│   ├── conftest.py             # Test fixtures
│   └── test_scraper.py         # Scraper tests
├── server.py                   # MCP server implementation
├── requirements.txt            # Python dependencies
├── pyproject.toml              # Project configuration (Ruff, Pytest)
├── alltrails_mcp_venv/         # Virtual environment
├── README.md                   # This file
└── .gitignore                  # Git ignore file

How It Works

MCP Protocol: Uses the Model Context Protocol to communicate with Claude Desktop
Web Scraping: Scrapes AllTrails website for trail data using BeautifulSoup
Data Processing: Formats and returns trail information in a structured format
Tool Integration: Exposes tools that Claude can call to search and retrieve trail data

License

MIT License

Acknowledgments

Built using the Model Context Protocol
Trail data sourced from AllTrails
Inspired by the MCP community examples

⚠️ IMPORTANT NOTICE

This project is DEPRECATED and non-functional as of January 25, 2026.

The scraping functionality violated AllTrails' Terms of Service and has been completely disabled at their request. This repository exists only for historical/educational purposes.

DO NOT attempt to use, fork, or replicate this code for scraping AllTrails or any other website without explicit written permission from the website owner.

Respect Terms of Service. Respect copyright. Use official APIs when available.