Mapflow MCP for AI. Extract Buildings, Roads, and Features from Imagery.

Q: How do I start analyzing an image with Mapflow? (Using createprocessing)

You call the createprocessing tool, passing in the JSON structure containing the image data and specifying which AI model you want to run. This action sends the job into the queue.

Q: I ran a job; how do I know if it's finished? (Using getprocessingstatus)

You check getprocessingstatus and provide the unique task ID. It tells you if the status is 'Queued', 'Running', or 'Completed'. You keep calling it until you see 'Completed'.

Q: What data formats does Mapflow give me? (Using getprocessingresult)

The getprocessingresult tool returns vector datasets, typically in standard formats like GeoJSON or Shapefile. These are ready for direct import into most GIS software.

Q: I need to run 10 different projects. What should I do? (Using listprojects)

You should use listprojects first. This shows you all your existing work containers, ensuring that when you start a new job via createprocessing, it attaches correctly to the right project.

Q: How do I check what kind of features Mapflow can detect using listmodels?

Run listmodels to see all available AI models. This returns a comprehensive catalog of detectors, including 'Building Footprints,' 'Road Networks,' and 'Forest Cover.' You use this output to decide which model you need before starting any processing job.

Q: Where can I find records of my past runs using listprocessings?

You use listprocessings to pull a history of all your geospatial jobs. This shows the unique ID, start date, and model used for every run—even if they weren't part of an active project.

Q: What specific structure does the input need when I use createprocessing?

The createprocessing tool requires a JSON string containing necessary parameters. You must include the source imagery ID, the target model name, and any regional bounding box coordinates to start analysis correctly.

Q: Should I always use createproject before running an analysis?

Yes, calling createproject first establishes a dedicated container for your work. This keeps all related tasks—the initial processing job, subsequent status checks, and final results—grouped together under one project name.

Claude

ChatGPT

Cursor

Gemini

Windsurf

VS Code

JetBrains

Vercel

See Vinkius in Action

Works with every AI agent you already use

…and any MCP-compatible client

Connect to your AI in seconds.

Mapflow connects your AI client to geospatial processing. It automatically extracts features—like buildings, roads, and vegetation—from satellite or drone imagery.

You initialize a project, trigger an AI model run, monitor its status, and pull back clean vector datasets (GeoJSON, Shapefile) ready for GIS analysis.

What your AI can do

Create processing

Starts a new job that analyzes satellite imagery based on your input data.

Create project

Sets up a container or folder to hold all related mapping projects and analyses.

Get processing result

Downloads the final, structured vector data once an analysis job has finished successfully.

+ 4 more capabilities included

Create New Mapping Projects

Initializes a container for related geospatial tasks, keeping your analyses organized by project name.

Run Imagery Analysis Jobs

Triggers the AI models to process raw satellite or drone images, detecting specific features like buildings or roads.

Check Job Status and Results

Monitors active processing tasks, providing real-time status updates and retrieving finished vector datasets.

List Available AI Models

Retrieves a list of all specialized models (e.g., forest cover, building footprints) you can apply to your imagery.

Ask an AI about this

Included with Plan

Waiting for input…

AI Agent

Mapflow MCP Server: 7 Tools for Geospatial Analysis

These tools allow you to manage the entire lifecycle of an analysis—from creating a project scope to running complex AI models and downloading final vector data.

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 Mapflow on Vinkius

Create Processing

Starts a new job that analyzes satellite imagery based on your input data.

Create Project

Sets up a container or folder to hold all related mapping projects and analyses.

Get Processing Result

Downloads the final, structured vector data once an analysis job has finished...

Get Processing Status

Checks if a specific background processing task is running, queued, or complete.

List Models

Displays every specialized AI model available for feature extraction (e.g., roads...

List Processings

Shows a list of all background processing jobs you’ve run against the platform.

List Projects

Retrieves and displays all the mapping projects you have created with Mapflow.

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 Mapflow integration is available immediately — no restart needed.

Antigravity

Configure Agent Environment

Open your Antigravity agent's workspace configuration or mcp-servers.json file.

Bind the Endpoint

Add the Vinkius endpoint URL to your agent's MCP connections list:

"mcp_servers": {
  "mapflow": {
    "serverUrl": "https://edge.vinkius.com/[YOUR_TOKEN_HERE]/mcp"
  }
}

Provide your secure token in place of [YOUR_TOKEN_HERE] to ensure your agent requests are authenticated.

Execute

Start your Antigravity session. The agent will autonomously discover and utilize the Mapflow tools with full Vinkius guardrails applied.

VS Code Copilot

⚡

One-Click Install (Recommended)

In your Vinkius Dashboard, simply click the Add to VS Code button for this server. We'll automatically configure your local workspace.

Or configure manually

Open MCP Settings

Open VS Code, press Ctrl/Cmd + Shift + P, and search for GitHub Copilot: MCP Servers.

Add Server Config

Add the Vinkius endpoint configuration to your mcp-servers.json file:

"mapflow": {
  "url": "https://edge.vinkius.com/[YOUR_TOKEN_HERE]/mcp"
}

Ensure you replace [YOUR_TOKEN_HERE] with your token from cloud.vinkius.com.

LangChain

Install Dependencies

Install the LangChain MCP adapters for your environment:

pip install langchain-mcp-adapters

Connect the Server

Use the SSEClient in LangChain to connect to the Vinkius managed endpoint:

from langchain_mcp_adapters.client import SSEClient

# Connect to Vinkius
client = SSEClient(url="https://edge.vinkius.com/[YOUR_TOKEN_HERE]/mcp")
tools = client.get_tools()

CrewAI

Define the Tool

Load the Vinkius MCP tools into your CrewAI agents:

from crewai import Agent
from mcp_crewai import MCPTool

# Connect securely to Vinkius
vinkius_tools = MCPTool(url="https://edge.vinkius.com/[YOUR_TOKEN_HERE]/mcp")

# Assign to Agent
researcher = Agent(
    role='Data Researcher',
    tools=vinkius_tools.get_all()
)

Execute Task

Run your CrewAI process. The agent will autonomously route tasks to the Vinkius managed server.

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

Start building

Make Your AI Do More

Start with Mapflow, 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 Mapflow. 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

Your data is protected. See how we built it.

Works with 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 7 powerful capabilities that interface natively with Claude, ChatGPT, Cursor, and other compatible AI platforms. No middleware. No custom integration required.

Getting usable geospatial data used to require painstaking manual work.

Before Mapflow, if you needed to map infrastructure changes—say, counting every building in a new sector—you were stuck. You'd download massive TIFF files and then spend days running semi-automated scripts or manually tracing boundaries in GIS software. It was slow, error-prone, and expensive.

Now, your agent calls `create_processing` with the image and selects 'Building Footprints'. The AI handles the detection; you just check `get_processing_status` until it’s done. When you get to `get_processing_result`, you're handed a clean dataset of coordinates—no manual tracing required.

Mapflow MCP Server: Extracting features from imagery.

Previously, managing a multi-stage analysis was a mess. You’d initiate the job in one place and then have to switch interfaces—a dashboard here, a command line there—to check if it finished or grab the data. The entire process lacked coherence.

With Mapflow, you keep the whole workflow inside your agent client. You define the project using `create_project`, trigger the work with `create_processing`, and retrieve everything from one place. It keeps your complex analysis entirely contained.

Support 24/7 support@vinkius.com ↗

Security Vinkius Trust Center ↗

SLA Service Level Agreement ↗

Report Listing Send Report ↗

What your AI can actually do with this

Mapflow connects your AI client to geospatial processing. It lets you automatically pull features—like buildings, roads, and vegetation—right out of satellite or drone imagery without writing a single line of code. You'll use this server to manage complex mapping projects and run deep-learning models on huge image files.

Setting Up Your Workspace:

You start by keeping all your analyses neat and separate using create_project. This tool sets up a dedicated folder, or container, for every related geospatial task you’re doing. You'll use list_projects to pull up a list of every single mapping project you've ever set up with Mapflow. If you need to see what projects already exist, just call list_projects.

Selecting Models and Jobs:

Before you run anything, you gotta know what AI models are out there. You can use list_models to pull a list of every specialized model available for feature extraction. These include specific tools for identifying things like forest cover or building footprints. Once you've chosen your scope, you initiate the work by calling create_processing.

This tool takes your raw image data and lets you specify the exact AI model it should use to start analyzing the imagery.

Tracking the Work:

Processing big images takes time, so you need a way to track what's going on. You can check if a specific background processing task is running, queued up, or finished by using get_processing_status. If you want a full history of all the jobs you've run against the platform, you use list_processings to pull that list.

When you need to manage your projects themselves, calling list_projects also shows you what containers are active.

Getting the Results:

The moment the job is finished, you can grab the structured data using get_processing_result. This tool downloads the final vector datasets—think GeoJSON or Shapefile formats—ready for immediate GIS analysis. Because you've got a running list of all your jobs via list_processings, you know exactly which processing run corresponds to the results you need to download.

The Workflow in Action:

You first establish order by calling create_project to set up a container for your specific mapping scope. Then, using that project context, you kick off the analysis with create_processing, feeding it the image data and telling it which AI model—like 'Road Networks' or 'Commercial Structures'—to use. You’ll then monitor progress by checking the status with get_processing_status; if you need to review previous runs, list_processings gives you that history.

When the job is complete, you pull back the clean data using get_processing_result. Throughout this process, list_models lets you see every available AI tool, and list_projects keeps track of all your current and past work.

Built · Hosted · Managed by Vinkius Mapflow MCP Server - Geospatial Feature Extraction

Server ID 019dd120-14d2-732a-8078-fca794369339

Vinkius Inspector

Compliance Grade A+

Score 100/100

Report View Report ↗

Here's how it actually works

The bottom line is you manage a multi-step asynchronous workflow: set up > trigger > wait > download.

First, use create_project to define the scope and name for your analysis.

Next, call create_processing, passing the image data and selecting a model via an existing project ID. This starts the job in the background.

Finally, periodically check the status using get_processing_status. When it returns 'Completed', use get_processing_result to get the final vector dataset.

Who is this actually for?

This tool is for GIS Specialists, Urban Planners, and Environmental Scientists. You need to analyze massive amounts of satellite imagery quickly without hiring an army of data labelers. If you're tired of manually tracing roads or counting buildings from maps, this saves days of grunt work.

Urban Planner

Uses create_processing with 'Building Footprints' models to map new construction sites and assess urban density across a city block.

GIS Specialist

Runs batch jobs using list_models and create_project, then uses the seven tools to systematically extract vector data (polygons, lines) from multiple datasets for comparison.

Environmental Consultant

Applies models like 'Forest Cover' across large land tracts via a new project to monitor deforestation or track agricultural field boundaries over time.

What Changes When You Connect

Gets rid of manual feature counting. Instead of manually tracing boundaries on a map, you run create_processing with an AI model and get thousands of clean polygons back instantly.

Keeps your work organized across complex projects. Use list_projects and create_project to silo different analyses—like 'Seattle Buildings' vs. 'Amazon Forests'—so they never mix up data streams.

Knows exactly where the job stands. The get_processing_status tool lets you poll a task ID without guessing if it stalled or finished, saving time waiting on unknown processes.

Works with specialized models for specific needs. You don’t just get 'features'; you specify 'Forest Cover' using list_models, ensuring the AI uses the right detection method.

Delivers usable data formats. When a job is done, get_processing_result gives you GeoJSON or Shapefile—data ready to drop straight into ArcGIS or QGIS.

See it in action

01 01

Assessing Urban Expansion

An urban planner needs to know how many new structures appeared in a neighborhood. They use create_project for 'Downtown Sector', then run create_processing with the 'Building Footprints' model on recent drone imagery. Once done, they pull the results via get_processing_result and overlay them on old census data to calculate growth rates.

02 02

Monitoring Wildfires

An environmentalist tracks a burned area by running multiple jobs. First, they use list_models to select 'Forest Cover' and run it across the entire region via create_processing. They then check status with get_processing_status until all tiles are complete, giving them a full dataset of cleared land.

03 03

Infrastructure Audits

A civil engineer needs to map every major road intersection. They create a new project and use create_processing with the 'Road Networks' model across high-resolution satellite imagery, making it fast and accurate compared to manual mapping.

04 04

Comparative Analysis

A researcher wants to compare agricultural yield changes over two years. They set up one project and run create_processing twice—once for Year 1 data and once for Year 2 data, using the 'Agriculture Fields' model. This allows them to use get_processing_result on both sets of polygons side-by-side.

The honest tradeoffs

Treating it like a single call

Anti-pattern

Calling create_processing and then immediately calling get_processing_result. The job hasn't finished, so you get an error or incomplete data.

The Fix

You must use create_project first. Then, start the job with create_processing. After that, you must poll the status using get_processing_status repeatedly until it reports 'Completed' before attempting to call get_processing_result.

Forgetting available models

Anti-pattern

Just running a generic analysis and hoping for the best. You might extract features, but they won't be standardized or machine-readable.

The Fix

Always start by using list_models. This shows you precisely which AI models are specialized (like 'Road Networks') and ensures your job uses the correct detection mechanism.

Losing track of jobs

Anti-pattern

Running five different analyses and then not knowing which one is finished or where to find the data. Your workflow stalls.

The Fix

Use list_processings right away. It gives you a master list of all your job IDs, letting you easily reference them when calling get_processing_status for tracking.

When It Fits, When It Doesn't

Use this server if your core task involves moving from raw image data (satellite/drone) to structured, clean vector data (polygons, lines). You need a repeatable workflow: setup -> trigger -> wait -> download. Don't use it if you only need simple metadata lookup or basic text analysis—use a general document processing tool instead. If your goal is just to see what models exist, use list_models first. If you are managing multiple separate analyses over time, start by checking list_projects. The workflow always begins with defining the scope (create_project) and ends with retrieving the structured output (get_processing_result).

Questions you might have

How do I start analyzing an image with Mapflow? (Using create_processing) +

You call the create_processing tool, passing in the JSON structure containing the image data and specifying which AI model you want to run. This action sends the job into the queue.

I ran a job; how do I know if it's finished? (Using get_processing_status) +

You check get_processing_status and provide the unique task ID. It tells you if the status is 'Queued', 'Running', or 'Completed'. You keep calling it until you see 'Completed'.

What data formats does Mapflow give me? (Using get_processing_result) +

The get_processing_result tool returns vector datasets, typically in standard formats like GeoJSON or Shapefile. These are ready for direct import into most GIS software.

I need to run 10 different projects. What should I do? (Using list_projects) +

You should use list_projects first. This shows you all your existing work containers, ensuring that when you start a new job via create_processing, it attaches correctly to the right project.

How do I check what kind of features Mapflow can detect using list_models? +

Run list_models to see all available AI models. This returns a comprehensive catalog of detectors, including 'Building Footprints,' 'Road Networks,' and 'Forest Cover.' You use this output to decide which model you need before starting any processing job.

Where can I find records of my past runs using list_processings? +

You use list_processings to pull a history of all your geospatial jobs. This shows the unique ID, start date, and model used for every run—even if they weren't part of an active project.

What specific structure does the input need when I use create_processing? +

The create_processing tool requires a JSON string containing necessary parameters. You must include the source imagery ID, the target model name, and any regional bounding box coordinates to start analysis correctly.

Should I always use create_project before running an analysis? +

Yes, calling create_project first establishes a dedicated container for your work. This keeps all related tasks—the initial processing job, subsequent status checks, and final results—grouped together under one project name.

Can I process satellite imagery and extract vector data? +

Yes. Trigger AI models on imagery to automatically extract buildings, roads, forests, and other features as vector polygons.

How does Mapflow authentication work? +

Mapflow uses Bearer authentication against api.mapflow.ai/rest using your API Key.

Can I track processing tasks? +

Yes. Monitor processing status, progress percentages, and access completion datasets when ready.

Connect to your AI in seconds.

Create processing

Create project

Get processing result

Mapflow MCP Server: 7 Tools for Geospatial Analysis

Make your AI actually useful.

Create Processing

Create Project

Get Processing Result

Get Processing Status

List Models

List Processings

List Projects

Security and governance baked right in.

Claude AI

Open Claude Settings

Add Custom Connector

Start a conversation

Claude Code

Open your terminal

Add the MCP Server

Start coding

Cursor

One-Click Install (Recommended)

Open Cursor Settings

Add New Server

Use in Composer

Antigravity

Configure Agent Environment

Bind the Endpoint

Execute

VS Code Copilot

One-Click Install (Recommended)

Open MCP Settings

Add Server Config

Windsurf

One-Click Install (Recommended)

Open Windsurf Settings

Add Server Endpoint

LangChain

Install Dependencies

Connect the Server

CrewAI

Define the Tool

Execute Task

Choose How to Get Started

Build Your Own

Make Your AI Do More

Works with Claude, ChatGPT, Cursor, and more

Getting usable geospatial data used to require painstaking manual work.

Mapflow MCP Server: Extracting features from imagery.

What your AI can actually do with this

Here's how it actually works

Who is this actually for?

What Changes When You Connect

See it in action

Assessing Urban Expansion

Monitoring Wildfires

Infrastructure Audits

Comparative Analysis

The honest tradeoffs

Treating it like a single call

Forgetting available models

Losing track of jobs

When It Fits, When It Doesn't

Questions you might have