alltrails

This MCP server historically exposed AllTrails data through the MCP protocol, allowing clients to perform two main tools: search_trails and get_trail_details. Clients could ask the server to search for trails within a specific national park using a park slug like us/tennessee/great-smoky-mountains-national-park, or request detailed information about a particular trail by providing the AllTrails slug. The server communicated via standard MCP protocol (stdio) and supported MCP 1.9.4, enabling natural language style queries and structured tool calls to retrieve trail data. Note that this project is deprecated and no longer functional, but the tooling model provides a reference for how MCP-driven data access can be exposed for trail information.

To use it, connect your MCP client (or Claude Desktop-style integration) to the configured server endpoint. Invoke search_trails with a park slug to retrieve matching trails, then optionally call get_trail_details with a trail slug to obtain details such as difficulty, length, and elevation gain. Since the project is historical, expect that live AllTrails data scraping is disabled and the server may not respond as originally intended.

Prerequisites:

• Python 3.10+ (as indicated by the project) or a compatible Python environment
• Access to the project repository files (cloneable)

Option A: Install and run locally (recommended for historical/reference)

1. Clone the repository

git clone <your-repo-url>
cd alltrails_mcp

2. (Optional) Create and activate a virtual environment

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

3. Install dependencies

pip install -r requirements.txt

4. Run the MCP server

python3 server.py

Option B: Docker (historical guidance)

1. Build the Docker image

docker build -t alltrails-mcp .

2. Run the container

docker run -i --rm alltrails-mcp

Option C: Claude Desktop / MCP client configuration (historical examples)

• Configure an MCP server entry with command python3 and arguments ["server.py"], or with a virtual environment path to python and the server script as arguments.

Prerequisites specific to your setup (e.g., virtualenv activation, paths) should be adapted to your environment.

Important: The AllTrails MCP server described here is deprecated and no longer functional due to changes in AllTrails' terms and scraping restrictions. Do not rely on it for live data. If you need trail data, use AllTrails' official site or API access where permitted, and ensure compliance with terms of service. If you attempt to run this historically, you may need to adjust Python paths, install-time dependencies, and ensure the server script (server.py) is present in your working directory. The MCP protocol version used was 1.9.4 with stdio communication. If integrating with Claude Desktop or other MCP clients, ensure their configuration matches the path to your server script and the correct Python interpreter.