Mapbox Alternative MCP for AI. Calculate routes, query layers, and build maps with simple prompts.
Gemini Works with every AI agent you already use
…and any MCP-compatible client
How this MCP server connects to your AI agent
Mapbox Alternative lets your AI agent handle complex geospatial tasks: calculate driving routes, run spatial queries on datasets, and generate static map images using natural language prompts.
You can use it for geocoding addresses, finding optimal travel paths between multiple points, or viewing data layers directly from a chat window.
What AI agents can do with Mapbox Automation
Geocode reverse
Takes coordinates and figures out what address or place they belong to.
Get dataset feature
Pulls specific data points (features) from a defined dataset by ID.
Get directions
Calculates the shortest path, providing turn-by-turn instructions and travel time estimates.
Calculates turn-by-turn directions and the best travel time between two or more points.
Takes a set of coordinates (lat/lon) and returns human-readable addresses using forward geocoding.
Performs interactive location searches, providing autocomplete suggestions for addresses or points of interest.
Lets you list, retrieve details on, and update specific geographic features within your linked datasets.
Calculates the geographical area a point can reach given a specified time limit or distance radius.
Ask an AI about this
Waiting for input…
What AI agents can do with Mapbox Alternative: 27 Tools for Geospatial Operations
These tools give your AI client direct access to every Mapbox API function—from basic geocoding and searching to advanced optimization and data layer management.
Make your AI actually useful.
Add this MCP to Claude, Cursor, or Windsurf and your AI stops guessing. It gets real tools to look things up, take action, and handle the stuff you keep doing by hand.
Start using Mapbox on VinkiusGeocode Reverse
Takes coordinates and figures out what address or place they belong to.
Get Dataset Feature
Pulls specific data points (features) from a defined dataset by ID.
Get Directions
Calculates the shortest path, providing turn-by-turn instructions and travel time...
Get Feedback
Collects user feedback about your mapping experience.
Get Isochrone
Computes a shape showing all points reachable from a starting point within a...
Get Map Matching
Snaps rough GPS tracks to the nearest valid road network lines for cleanup.
Get Matrix
Returns travel times and distances between multiple different points simultaneously.
Get Optimization
Recovers a solution that solves complex routing problems, like the Traveling...
Get Raster Tile
Downloads an image tile (PNG, JPG) from a Mapbox-hosted tileset.
Create Tileset Source
Builds a source for the Mapbox Tiling Service (MTS).
Create Token
Generates and stores a new access token for your account.
Delete Token
Removes an existing access token from your profile.
Geocode Batch
Converts many addresses into coordinates in one batch process.
Geocode Forward
Finds the nearest address or location based on a set of coordinates.
Get Static Image
Generates a single, viewable map image for specific coordinates and zoom level.
Get Tilequery
Retrieves detailed data about features at a precise point on the map (lat/lon).
Get Vector Tile
Downloads raw, structured vector data (MVT) from a Mapbox-hosted tileset.
List Datasets
Lists all geographic datasets associated with your account.
List Tokens
Shows you every access token currently stored on your profile.
Publish Tileset
Makes a Mapbox Tiling Service (MTS) tileset publicly available for others to use.
Searchbox Forward
Performs an initial search lookup using the Search Box API.
Searchbox Retrieve
Gets full, detailed information about a location suggested by the user.
Searchbox Reverse
Performs a reverse search to find a known place by coordinates.
Searchbox Suggest
Provides real-time autocomplete suggestions as you type an address or location name.
Submit Feedback
Allows you to submit general user feedback regarding the service usage.
Submit Optimization
Runs a complex algorithm to solve multi-stop route planning problems for fleets.
Update Dataset Feature
Changes or corrects specific data points within an existing dataset feature.
Security and governance baked right in.
Pick your AI client below to get set up. Just create a Vinkius account, subscribe, and you're instantly up and running. We handle the entire backend infrastructure, delivering out-of-the-box support for HTTPS Streamable, SSE, and OAuth2—zero messy routing required.
Claude AI
Open Claude Settings
Go to claude.ai, click your profile icon, then navigate to Customize → Connectors.
Add Custom Connector
Click the "+" button and select Add custom connector. Paste your Vinkius endpoint URL:
https://edge.vinkius.com/[YOUR_TOKEN_HERE]/mcp
Replace [YOUR_TOKEN_HERE] with your token
from cloud.vinkius.com. For OAuth-protected servers, expand
Advanced settings to add credentials.
Start a conversation
Open a new chat. The Mapbox Alternative integration is available immediately — no restart needed.
Choose How to Get Started
Build a custom MCP for your own tools, or connect a ready-made integration from our catalog.
Build Your Own
Turn any API into an MCP. Import a spec, define Agent Skills, or deploy with MCPFusion.
- Import from OpenAPI, Swagger, or YAML specs
- Create Agent Skills with progressive disclosure
- Deploy to edge with MCPFusion framework
- Built in DLP, auth, and compliance on every call
- Real time usage dashboard and cost metering
- Publish to catalog or keep private
Make Your AI Do More
Start with Mapbox, then connect any of our 5,100+ other servers whenever your AI needs more. One click, no limits.
- Use this MCP plus 5,100+ others, all in one place
- Add new capabilities to your AI anytime you want
- Every connection is secured and compliant automatically
- Track usage and costs across all your servers
- Works with Claude, ChatGPT, Cursor, and more
- New servers added to the catalog every week
Independent Platform Disclaimer: Vinkius is an independent platform and is not affiliated with, endorsed by, sponsored by, verified by, or otherwise authorized by Mapbox. All third-party trademarks, logos, and brand names are the property of their respective owners. Their use on this website is strictly for informational purposes to identify service compatibility and interoperability.
VINKIUS INFRASTRUCTURE
Cloud Hosted
Managed infra
V8 Isolated
Sandboxed per request
Zero-Trust Proxy
No stored credentials
DLP Enforced
Policy on every call
GDPR Compliant
EU data residency
Token Compression
~60% cost reduction
Built on the Model Context Protocol (MCP) for Claude, ChatGPT, Cursor, and more
The Model Context Protocol standardizes how applications expose capabilities to LLMs. Instead of operating in isolation, your AI gains direct access to external platforms, live data, and real-world actions through secure, standardized connections.
This connection provides 27 powerful capabilities that interface natively with Claude, ChatGPT, Cursor, and other compatible AI platforms. No middleware. No custom integration required.
Mapping out service areas used to take an hour of tab-switching., Solved with Vinkius AI Gateway
Today, if you need to know the maximum area your new branch can cover in 25 minutes, you'd open a GIS platform. You'd input coordinates, adjust time sliders, run simulations, and then export a PDF showing the resulting shape. It's slow, it requires specialized software, and it’s hard to iterate on.
With this MCP server, you just prompt your agent: 'Show me all locations reachable from 123 Main St in under 25 minutes.' The agent runs `get_isochrone` instantly. You get a clean, actionable shape returned right in the chat. No software needed.
Using Mapbox Alternative MCP Server: Get map data and routes in one prompt.
Before this, calculating multi-stop routes meant opening a spreadsheet, manually listing every coordinate, and then pasting that massive list into an external routing engine. If you changed the order of two stops, you had to restart the whole process.
Now, you tell your agent: 'Calculate the optimal sequence for these 12 delivery points.' The agent runs `get_matrix` first, gathering all distances, and then uses `submit_optimization`. It gives you the solution in seconds. That's a huge difference.
What your AI can actually do with this
Listen up. This server lets your AI client handle heavy-duty mapping and geospatial work right from chat. You don't need a whole separate stack of services; you just talk to your agent, and it handles the coordinates, routes, and data pulls for ya.
Finding Locations (Search & Geocoding)
You can nail down any address or point of interest using interactive searches. When you start typing a location name, the system uses searchbox_suggest to give you real-time autocomplete options. Once you pick one, running searchbox_forward performs an initial search lookup for that spot. If you need all the details—like the full address and geometry—you use searchbox_retrieve.
Need to work backward? Use geocode_reverse; it takes coordinates (lat/lon) and tells you what street address they belong to. For the opposite, if you have an address string but need the exact numbers, run geocode_forward to find the nearest valid location based on your set of coordinates.
If you're dealing with a bunch of addresses at once—say, a whole list for a mailing campaign—you batch them up using geocode_batch, and it converts all those names into coordinates in one go. To check what specific address a set of coordinates is closest to, you use geocode_forward again.
Calculating Routes & Movement (The Hard Stuff)
This is where the power's at. You can calculate turn-by-turn driving directions and get time estimates using get_directions. If you need to know how long it takes to get between multiple points, or what the distance is for a bunch of different spots simultaneously, run get_matrix.
For complex trips—like having a delivery driver hit five different stops and needing the absolute shortest path—you use get_optimization. You can also feed those multi-stop problems into submit_optimization to solve them for your whole fleet. Beyond simple driving, you can figure out areas reachable within a specific time or distance using get_isochrone, which computes that required shape.
Working with Data & Maps (Datasets and Tiles)
You don't just get lines on a map; you get data. You first check what datasets are available to you by calling list_datasets. Once you know the dataset, you can pull specific points of interest or data features using get_dataset_feature, and if that feature needs correcting, run update_dataset_feature.
When you need to pinpoint exactly what kind of data lives at a precise spot on the map (a latitude and longitude), use get_tilequery. If you're downloading raw structured geometry for advanced processing, get_vector_tile gives you that Mapbox Vector Tile (MVT) format. For standard images, get_raster_tile downloads a basic image tile (PNG or JPG).
You can also create and manage custom map layers: use create_tileset_source to build your source for the Tiling Service, then you can make it public with publish_tileset. To generate a single, viewable JPEG of a specific area at a given zoom level, run get_static_image.
Utility and Account Management
To get started or manage access, the system lets you handle tokens. You can create a new secure token using create_token, see every token stored on your profile with list_tokens, delete old ones with delete_token, and review all of that history by calling list_tokens. If you want to pass along any thoughts about how the service works, use get_feedback or submit_feedback.
019e5d32-cbc6-70fd-a25d-f6d50c920287 
Here's how it actually works
The bottom line is you send a simple request in natural language and get actionable geographical data back without leaving your editor.
Subscribe to this server and enter your Mapbox Access Token (Public or Secret).
Your AI client uses the token to connect to the geospatial API.
You prompt the agent with a query, like 'Calculate the fastest route from A to B'—the system returns the structured data.
Who is this actually for?
Anyone who deals with geography or logistics needs this. If you're the operations manager tired of manually calculating complex delivery routes, or the GIS analyst who wastes time switching between coordinate systems, this is for you. It puts all that math and mapping power right into your chat window.
Uses get_matrix and submit_optimization to calculate the most efficient sequence of stops for a delivery fleet.
Runs tools like list_datasets and get_tilequery to inspect raw feature data and check tileset statuses without touching a terminal.
Uses geocoding and spatial queries (get_isochrone) to map out potential service areas or analyze geographic trends for reports.
What Changes When You Connect
Stop copying coordinates. Use geocode_forward or searchbox_suggest to let your agent handle all lookups, converting addresses into precise lat/lon points instantly.
Don't waste time on single routes. Run multi-stop logistics using get_matrix to compare travel times and distances between dozens of locations in one call. Then refine the order with submit_optimization.
Need visual data? Instead of opening a map editor, use get_static_image to generate an image right in your chat window for reports or presentations.
Working with raw GIS data? Use list_datasets and then get_dataset_feature to pull specific attributes (like population density or zoning codes) directly from the dataset structure.
Planning service coverage? Don't guess. Run get_isochrone to precisely calculate the area a branch can cover within, say, 20 minutes of driving time.
See it in action
Optimizing a Delivery Route
A field manager has 15 stops and needs the most efficient order. Instead of mapping it out manually in an external tool, they prompt their agent: 'Calculate the optimal route for these 15 addresses.' The agent runs get_matrix to get all distances and then uses submit_optimization, returning the perfect, sequenced driving plan.
Checking Service Area Boundaries
A retail buyer needs to know if a new store location is within reasonable reach of their existing 10 distribution centers. They ask the agent to run get_isochrone from the proposed site, using a 30-minute driving radius to check immediate coverage.
Analyzing Data Layers in GIS
A city planner needs to know if specific zoning rules apply at a parcel of land. They first use list_datasets to find the 'Zoning Map,' then run get_tilequery on the coordinates, getting back the exact regulatory data needed for their report.
Quickly Generating Presentation Maps
A real estate agent needs a map of a neighborhood's potential selling area for a client pitch. They ask the agent to 'Generate a static image of this block at zoom 15.' The agent uses get_static_image and returns a ready-to-paste, high-quality JPEG.
The honest tradeoffs
Copy-pasting coordinates for everything
Trying to manually run a series of lookups: first geocode_forward for Point A, then using those results in another tool's parameters. This is slow and error-prone.
Use the agent to manage the flow. Start with searchbox_suggest to nail down the location name, which gives you coordinates. Then pass those coordinates directly into a complex calculation like get_matrix or get_isochrone.
Ignoring the dataset structure
Assuming that all map data is always available through simple geocoding. You might miss specific metadata because you didn't check what layers exist.
Always start by running list_datasets. This shows exactly which specialized datasets are connected to your account, letting you target the right tool like get_dataset_feature.
Treating directions as a single step
Just asking 'How do I get there?' and getting only basic directions. You miss out on optimizing for multiple stops.
For anything involving more than two points, use get_matrix first to collect all travel times/distances. Then pass that list into submit_optimization to solve the routing puzzle.
When It Fits, When It Doesn't
Use this server if your problem has a clear geographic component: finding a path, determining a service area, or reading data from a map layer. Don't use it if you just need simple text formatting, document summarization, or general database CRUD operations that don't involve location (e.g., updating a user's name field). If your job is purely textual and doesn't care about coordinates, stick to standard LLM functions instead of calling get_directions.
However, if you are dealing with complexity—like optimizing 20 stops or calculating reachability across different time limits—you need the full suite. The core difference between simple lookups (geocode_forward) and advanced tools like get_matrix is that the latter requires multiple inputs and solves a mathematical problem (minimum distance/time) rather than just finding a single point.
Questions you might have
How do I use get_directions to find a route? +
You ask your agent for directions between two points, providing coordinates or addresses. The agent uses get_directions and returns the optimal path, including turn-by-turn instructions and total travel time.
What is the difference between get_matrix and get_isochrone? +
The key difference is purpose: get_isochrone calculates a reachable area (a shape) from one point based on time/distance. get_matrix calculates multiple specific travel times and distances between many distinct points.
Can I update data using get_dataset_feature? +
Yes, if you know the feature ID and the new data values, you can run update_dataset_feature. This lets your agent change a specific attribute (like changing a business's operating hours) within your connected dataset.
How do I search for an address using searchbox_suggest? +
You simply ask the agent to 'Find restaurants near 123 Main St.' The agent uses searchbox_suggest and provides live, autocomplete suggestions as you refine the location name.
What should I use for authentication when calling `get_dataset_feature`? +
You must pass an active access token. Use the create_token tool first to generate a secure key, then include that token in your agent's context. This keeps your data interactions authenticated and traceable.
When should I use `get_vector_tile` instead of `get_raster_tile`? +
Vector tiles provide raw, structured geometry (MVT or PBF). Raster tiles deliver pre-rendered images (PNG/JPG). Use vector when you need to style the map dynamically within a client; use raster for quick visual output.
How does `submit_optimization` solve complex fleet routes? +
It solves the Traveling Salesperson Problem (TSP). You feed it multiple required stops, and the tool returns the optimal sequence to hit every location. It minimizes total travel time for your entire route.
What's the most efficient way to geocode hundreds of addresses? +
Use geocode_batch for bulk processing. Instead of running single-point searches repeatedly, this tool handles large lists of coordinates or addresses in one go. This saves calls and processes your data much faster.
Can I calculate driving directions between two points using this server? +
Yes! Use the get_directions tool. You need to provide a routing profile (like mapbox/driving) and the coordinates as a semicolon-separated string of longitude and latitude.
How do I find information about map features at a specific coordinate? +
You can use the get_tilequery tool. By providing a tileset ID and the longitude/latitude, the agent will retrieve data about the specific features located at that point.
Is it possible to update my Mapbox datasets through the AI? +
Yes, the update_dataset_feature tool allows you to modify a specific feature within a dataset by providing the dataset ID, feature ID, and the new GeoJSON data.
We've already built the connector for Mapbox Alternative. Just plug in your AI agents and start using Vinkius.
No hosting. No infrastructure. No complex setup.
All 27 tools are live and waiting.
You're up and running in seconds.
Gemini Vinkius gives your AI agents access to the full catalog of app connectors, all fully managed, secure, and enterprise-ready. One subscription, every tool you need.
Built, hosted, and secured by Vinkius. You just connect and go.