SmartRoutes MCP Server
Enables AI assistants to interact with SmartRoutes route optimization and dispatch tools, allowing users to manage orders, vehicles, and plans via natural language.
README
SmartRoutes MCP Server
Expose your SmartRoutes route-optimisation and dispatch tools to AI assistants like Claude Desktop, Claude Code, Cursor, Windsurf, and any other Model Context Protocol client — working against your live data.
Overview
The MCP server connects an AI tool to your live SmartRoutes account. Ask your AI Assistant which orders are still unscheduled, review a route plan, or kick off an optimisation — all in chat, against your real data.
It exposes the same operations as the SmartRoutes REST API, shaped as clean tools an agent can call directly. Everything is scoped to a single account by your API key.
Available Tools
| Tool | Type | Description |
|---|---|---|
list_orders |
Read | Orders for the depot (filter by date/status; paginated) |
get_order |
Read | Full detail for a single order |
add_order |
Write | Create a new order (pickup, delivery, or shipment) |
check_booking_availability |
Read | Whether a time window + address can be served |
optimise_plan_for_date |
Write | Trigger route optimisation for a date |
get_optimisation_status |
Read | Poll an optimisation job until it finishes |
get_plan |
Read | Plan-level totals and route summary for a date |
get_route |
Read | Per-stop detail for a single route |
list_vehicles |
Read | Fleet vehicles with shifts, capacities, and skills |
get_vehicle |
Read | Full detail for a single vehicle |
list_customers |
Read | Customers for the depot |
Write tools come with a confirmation step. The assistant explains what it's about to do and waits for your approval — especially optimise_plan_for_date with auto-dispatch, which publishes the plan to drivers and notifies customers.
Setup
1. Get your API key
The MCP server uses your existing SmartRoutes API key — the same x-access-key you already use for the REST API.
To generate an API Key, login to your SmartRoutes account and go to Settings > Integrations > SmartRoutes Open API and click on the Generate API Key button.
Treat the key as a secret; it grants access to your account's data.
2. Connect your AI tool
Ready-to-copy config files for every client live in
examples/— drop in your key and go.
Quick start (Claude Code)
This repo includes a pre-configured .mcp.json. Just replace <your-api-key>:
{
"mcpServers": {
"smartroutes": {
"type": "http",
"url": "https://mcp.smartroutes.io/mcp",
"headers": {
"x-access-key": "<your-api-key>"
}
}
}
}
Claude Code auto-discovers this file when you open the project.
Claude Desktop
Add a Custom Connector for SmartRoutes by going to: Settings → Connectors → Add custom connector
Here you can add the following details:
- Name: SmartRoutes (or whatever you like)
- URL: https://mcp.smartroutes.io/mcp
Add your authentication header under Advanced settings / Request headers in that same dialog:
- Key: x-access-key
- Value: your actual API key
Alternatively, you can add the following to your Claude Desktop config (claude_desktop_config.json):
{
"mcpServers": {
"smartroutes": {
"url": "https://mcp.smartroutes.io/mcp",
"headers": {
"x-access-key": "<your-api-key>"
}
}
}
}
Cursor
In Cursor settings, add an MCP server:
- Name: smartroutes
- Type: Streamable HTTP
- URL:
https://mcp.smartroutes.io/mcp - Headers:
x-access-key: <your-api-key>
Windsurf
Add to your Windsurf MCP config (mcp_config.json):
{
"mcpServers": {
"smartroutes": {
"serverUrl": "https://mcp.smartroutes.io/mcp",
"headers": {
"x-access-key": "<your-api-key>"
}
}
}
}
Any MCP Client (generic)
The server uses Streamable HTTP transport:
POST https://mcp.smartroutes.io/mcp
x-access-key: <your-api-key>
Any MCP client that speaks Streamable HTTP can connect. Legacy SSE clients are handled automatically.
3. Local development
Against a locally-running server the endpoint is (use whatever port your local app listens on):
http://localhost:3000/mcp-api
Example Conversations
Once connected, you can ask your assistant things like:
- "What orders for tomorrow aren't on a route yet?"
- "Show me available vehicles and their shifts."
- "Can we fit a delivery to 12 Main St between 2–4pm on Friday?"
- "Optimise tomorrow's routes for all open orders — draft only, don't dispatch yet."
- "What's the status of that optimisation?"
- "Show me the plan for 2026-07-01 and the stops on route 3."
- "Add a 10 kg delivery for ACME at 5 King St."
API Key Management
Your MCP key is your SmartRoutes API key.
| Action | Description |
|---|---|
| Use | The same x-access-key as the REST API; scoped to one depot. |
| Disable | Disabling a key in your SmartRoutes account immediately blocks both MCP and REST access. |
Security
- Authentication: API key in the
x-access-keyheader. - Depot isolation: every key is scoped to a single depot, and all data is filtered by depot.
- Rate limiting: each tool has its own per-key rate limit (shared with the REST API's limits).
- Write confirmation: write tools (
add_order,optimise_plan_for_date) instruct the assistant to explain and confirm before running — especially auto-dispatch.
How it works
The MCP server runs the same engine as the SmartRoutes REST API — each tool calls the same underlying service, so MCP and the API always agree on behaviour and data. Transport is Streamable HTTP; authentication is your API key.
A typical "plan tomorrow's routes" flow:
list_orders/list_vehicles— see what's availablecheck_booking_availability— confirm a slot (optional)add_order— enter new workoptimise_plan_for_date— build the plan (confirm before auto-dispatch)get_optimisation_status— poll until it finishesget_plan/get_route— read the result
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.