Workflow Automation · Graphhopper + Claude
GraphHopper MCP with Claude: Route Planning and Delivery Optimisation
The short answer
Claude works with GraphHopper through the Model Context Protocol, an open standard that lets Claude securely use other software. Once connected with an API key, Claude can geocode addresses, plan routes across vehicle types, optimise delivery stops and check travel times. People set the permissions and approve consequential actions before anything changes.
What is Graphhopper?
GraphHopper is an enterprise-grade Directions API for routing, optimisation and geocoding. It works across several vehicle types, from cars and vans to bicycles and trucks. Logistics firms, delivery operators, field service teams, fleet managers and retailers use it to plan efficient journeys, convert addresses into map coordinates and estimate how long trips will take. It is built for businesses that need fast, dependable route planning at scale.
How Claude connects to Graphhopper through MCP
The Model Context Protocol is an open standard that acts like a universal adapter between Claude and other software. It lets Claude use a tool such as GraphHopper to read information and take actions, with permissions the organisation controls. Claude does not gain access to anything beyond what has been set up.
Once GraphHopper is connected, Claude can typically look up and convert addresses, plan routes for a chosen vehicle, sequence a set of delivery stops, and analyse travel times between locations. Staff can ask questions in ordinary language, such as which order of stops keeps a van's journey shortest, and receive a clear answer with the supporting figures.
Permissions and approvals keep people in charge. Access is limited to the functions each workflow needs, and the connection uses the API key as its sign-in method. Consequential actions, such as changing a dispatch plan or sending routes to drivers, can require a named person to approve them first. Every request can be logged for review.
What you can automate
Daily delivery route planning
Each morning, a dispatcher asks Claude to build routes for the day's orders for a given fleet of vans. Claude geocodes the delivery addresses, groups stops by area and proposes a sequence for each vehicle. The dispatcher reviews the plan, adjusts it if needed and approves it before drivers receive their routes.
Address validation and geocoding
Customer records often contain incomplete or inconsistent addresses. Claude can use GraphHopper to convert addresses into coordinates and flag entries that cannot be located. Operations teams then correct these records before orders are scheduled, which typically reduces failed deliveries and wasted driver time.
Field service visit scheduling
Engineers and technicians travel between customer sites throughout the day. Claude can estimate travel times between jobs and suggest an order of visits that keeps driving to a minimum. Schedulers can ask follow-up questions, such as what happens if a job is added late, and receive an updated view of the day.
Travel time analysis for planning
Operations managers often need to know how long journeys really take before committing to service promises. Claude can analyse travel times between regions or depots and summarise the results in plain language. This helps teams set realistic delivery windows and identify routes that consistently take longer than expected.
Fleet vehicle type comparison
Businesses that run mixed fleets want to know which vehicle suits each route. Claude can compare expected journey times and planning results for different vehicle types across the same set of stops. Fleet managers use these comparisons to decide where larger vehicles are needed and where smaller ones are more efficient.
An end-to-end workflow: Next-day delivery planning with order and driver systems
TriggerEach evening at a set time, new orders are confirmed in the order management system.
Claude pulls the confirmed delivery addresses for the next day from the order management system.
Claude uses GraphHopper to geocode each address and flag any that cannot be located.
Claude plans optimised routes across the available vans, respecting the vehicle types and stop limits set by the operations team.
A dispatch supervisor reviews the proposed routes and approves them or requests changes.
Approved routes and estimated arrival windows are written back to the scheduling system and shared with drivers and customers.
OutcomeDispatch teams spend less time building routes by hand, failed deliveries caused by bad addresses fall, and customers receive more accurate arrival windows. Results vary by operation, but these gains are typical of organisations that automate this step.
How we set it up
Free automation audit
We review how your team plans routes, geocodes addresses and schedules deliveries today. Together we identify the workflows where Claude and GraphHopper will save the most time, and agree what success looks like.
Secure API key connection
We connect Claude to GraphHopper using an API key, stored securely and limited to the access each workflow needs. Where a managed connector is more suitable, we set that up instead, always with least-privilege permissions.
Workflow design and integration
We design the end-to-end process across GraphHopper and your other systems, such as Microsoft 365, your CRM, ERP or ticketing platform. Staff approval points are built in for any action that changes a plan or reaches a customer or driver.
Testing against real cases
We test the workflow using past orders, addresses and routes from your own operation. We check accuracy, compare results with your current planning, and refine prompts and rules until the output meets your standards.
Launch, monitoring and training
We launch with monitoring and audit logs, so every request and approval can be reviewed. Your team receives hands-on training, and we provide support as the workflow is used at full scale.
Who benefits
- Logistics and dispatch managers
- Build and adjust daily routes in minutes, with clear reasons for each sequence of stops.
- Operations directors
- Gain reliable travel time data to set realistic delivery promises and control costs.
- Field service coordinators
- Re-plan visits quickly when jobs change, reducing unnecessary driving between sites.
- IT and data leads
- Keep access limited, logged and reviewable, with the API key managed under your own controls.
Security and control
- Access is limited to the GraphHopper functions each workflow needs, following least-privilege principles.
- The API key is stored securely and can be rotated or revoked at any time.
- People approve any action that changes a dispatch plan or reaches a driver or customer.
- Requests and outcomes are recorded in audit logs for review and compliance checks.
- Test results are reviewed against real cases before any workflow goes live.
For teams in the US, UK, India and the Middle East
Organisations in the United States and the United Kingdom can host the Claude workflow in a region that meets their data requirements. In India, we design the solution to support obligations under the Digital Personal Data Protection Act, 2023. For businesses in the Middle East, including the UAE and Saudi Arabia, we align the setup with local data protection laws. Hosting region is chosen to meet each client's specific requirements.
Learn more about Claude and MCP
Anthropic's AI model and assistant.
Secure links between Claude and business tools.
How engineers build and change software with Claude.
Hosted agents that run multi-step work.
The open standard Claude uses to connect to tools.
Frequently asked questions
What does the GraphHopper MCP server do?
The GraphHopper MCP server acts as a secure bridge between Claude and the GraphHopper Directions API. It lets Claude request geocoding, route planning, delivery stop optimisation and travel time analysis in response to plain-language questions. Claude only uses the functions that have been enabled for the workflow, and the results come back as clear, readable answers.
Is connecting GraphHopper to Claude secure?
Yes, when it is set up with care. The connection uses an API key, which is stored securely and limited to the required access. We apply least-privilege permissions, add approval steps for consequential actions and keep audit logs of every request. Your IT team keeps control of the key and can revoke it at any time.
What can Claude do with GraphHopper once it is connected?
Typically, Claude can convert addresses into coordinates, plan routes for different vehicle types, sequence delivery stops, and estimate travel times between locations. Staff can ask questions in everyday language and receive plans with supporting figures. Claude does not make changes to live dispatch plans without the approvals you define.
How long does setup take?
Timelines depend on the number of workflows and the systems involved. A focused single-workflow project often moves from audit to launch in a few weeks, while broader multi-system projects take longer. We confirm the timeline during the free automation audit, once we understand your operation.
Which plan or permissions do I need?
You need a GraphHopper account that issues an API key with the access your workflows require. The exact plan depends on your usage volumes and the functions you need. During the audit we review your current arrangement and advise on whether any changes to access are needed.