[!WARNING]
This MCP server uses thepolarsteps-api
package to access Polarsteps data through undocumented APIs. Important: Please read the associated legal disclaimer and terms of use before using this tool. By using this MCP server, you agree to the terms outlined in the API package and acknowledge the associated risks.
A Model Context Protocol (MCP) server that enables Claude and other AI assistants to access Polarsteps travel data. Query user profiles, trip details, travel statistics, and search through travel histories with natural language.
Until I add it to PyPI, the quickest way to get started is using Smithery:
npx -y @smithery/cli install @remuzel/polarsteps-mcp --client claude
Then configure your Polarsteps token.
You'll need your Polarsteps remember_token
to authenticate API requests.
remember_token
cookie for https://www.polarsteps.comSet your token as an environment variable:
export POLARSTEPS_REMEMBER_TOKEN="your_remember_token_here"
Add this configuration to your Claude Desktop settings:
{
"mcpServers": {
"polarsteps": {
"command": "uvx",
"args": ["--from", "git+https://github.com/remuzel/polarsteps-mcp", "polarsteps-mcp"],
"env": {
"POLARSTEPS_REMEMBER_TOKEN": "your_remember_token_here"
}
}
}
}
Once configured, you can ask Claude things like:
Test the MCP server locally with the inspector:
npx @modelcontextprotocol/inspector uvx --from git+https://github.com/remuzel/polarsteps-mcp polarsteps-mcp
For development or manual installation:
# Clone the repository
git clone https://github.com/remuzel/polarsteps-mcp
cd polarsteps-mcp
# Setup development environment
just setup
# or without just:
uv sync --dev && uv pip install -e .
just test
just test-mcp
just lint
Seamless access to top MCP servers powering the future of AI integration.