wunder-user-mcp
An MCP server that exposes the Wunder Mobility v2 end-user API as tools, enabling clients to browse vehicles, manage rentals, and create reservations on behalf of a signed-in user.
README
wunder-user-mcp
An MCP server that exposes the Wunder Mobility v2 end-user (User) API as tools. It lets an MCP client (Claude, etc.) browse vehicles, inspect and operate rentals, and create rentals/reservations on behalf of a single signed-in end user.
Authentication is handled entirely from a refresh token you provide via env var — the server exchanges it for short-lived access tokens and refreshes them automatically, so tool callers never deal with tokens.
Tools
| Tool | What it does | Underlying endpoint |
|---|---|---|
get_vehicles |
List available vehicles. With a latitude/longitude, annotates each vehicle with distance from the user and sorts ascending. |
GET /vehicles |
get_active_rental |
Return the user's ongoing rental(s) (ACTIVE / RESERVATION). |
GET /rentals |
rental_command |
Operate a rental: START, PARK, DRIVE, END, OPEN_TAILBOX, RENEW_RESERVATION, … |
POST /rentals/{id}/operation |
create_rental |
Create a rental in RESERVATION or ACTIVE state from a vehicle_id or vehicle_code. |
POST /rentals |
Configuration
All configuration is via environment variables (see .env.example):
| Variable | Required | Default | Purpose |
|---|---|---|---|
WUNDER_REFRESH_TOKEN |
✅ | — | Long-lived refresh token for the end user. |
WUNDER_TENANT |
✅ | — | Tenant short-code used in the API URL path. |
WUNDER_BASE_URL |
https://go.api.gourban.services/v1 |
API gateway base URL. | |
WUNDER_BRANCH_ID |
— | Default branch id for vehicle/rental calls. | |
WUNDER_DEFAULT_RADIUS_MI |
5 |
Radius (miles) used by get_vehicles when a location is given without a radius. |
|
WUNDER_HTTP_TIMEOUT |
30 |
Per-request timeout in seconds. |
Getting a refresh token
Sign in once via the Wunder Authentication API (e.g. POST /{tenant}/auth/sign-in-email,
/sign-in-phone-number, or /sign-in-api-client) and copy the refreshToken from the
response into WUNDER_REFRESH_TOKEN.
Run
Run straight from the GitHub repo — no clone and no PyPI publish required (uvx builds
from the Git source):
uvx --from git+https://github.com/slettmayer/wunder-user-mcp wunder-user-mcp
Pin a branch, tag, or commit for stability by appending @<ref>:
uvx --from git+https://github.com/slettmayer/wunder-user-mcp@main wunder-user-mcp
uvx caches the build, so to pick up new commits on a moving ref (e.g. @main) add
--refresh:
uvx --refresh --from git+https://github.com/slettmayer/wunder-user-mcp@main wunder-user-mcp
From a local checkout (development):
uv run wunder-user-mcp # editable install, reflects local edits
uvx --from . wunder-user-mcp # build from the current directory
In all cases WUNDER_REFRESH_TOKEN and WUNDER_TENANT must be set in the environment
(see Configuration). The server speaks MCP over stdio.
MCP client configuration
{
"mcpServers": {
"wunder-user": {
"command": "uvx",
"args": [
"--from",
"git+https://github.com/slettmayer/wunder-user-mcp",
"wunder-user-mcp"
],
"env": {
"WUNDER_REFRESH_TOKEN": "<refresh-token>",
"WUNDER_TENANT": "<tenant-code>",
"WUNDER_BASE_URL": "https://go-staging.api.gourban.services/v1",
"WUNDER_BRANCH_ID": "<optional-branch-id>"
}
}
}
}
WUNDER_BASE_URL defaults to production (https://go.api.gourban.services/v1); the
value above targets staging (go-staging). Set it to match your tenant's environment
— a staging tenant on the production host returns 404. To pin a ref, append @<ref> to
the URL, e.g. git+https://github.com/slettmayer/wunder-user-mcp@main.
Development
uv sync # install deps (incl. dev group)
uv run pytest # offline unit tests (geo + JWT exp parsing)
uv run ruff check # lint
Notes / caveats
- The vehicle position field path used for distance calculation is detected
best-effort (
lat/lng,latitude/longitude, or a nestedposition/location). If a tenant's response uses different field names, extendgeo.py. operation_typeinrental_commandis passed through to the API, so tenant-specific operations beyond the documented set keep working.- Rental creation and
ENDare high-impact: they can trigger vehicle commands, payment authorization/capture, deposits, and invoices.
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.