Felt (Collaborative Maps) MCP. Control every element on your shared map from conversation.
Gemini Works with every AI agent you already use
…and any MCP-compatible client
Just plug in your AI agents and start using Vinkius.
Felt (Collaborative Maps) connects your AI agent directly to collaborative mapping services. You can upload complex datasets like GeoJSON, CSV, and KML, build maps from scratch, and manipulate every point, line, and polygon on a shared map using just conversation.
It gives you full control over geospatial data without opening any desktop GIS software.
What your AI agents can do
Add elements
Adds new geometric features like points and lines to a specified layer.
Create layer
Uploads raw geographic data (CSV, KML, etc.) from a URL and builds it into a new map layer.
Create map
Sets up an entirely new Felt map project with specific viewing boundaries.
Retrieves a list of all accessible maps or provisions a brand new collaborative map instance.
Takes raw geographic data (CSV, KML, GeoJSON) from URLs and converts it into usable layers on the map.
Adds new points, lines, or polygons to an existing layer using coordinates or defined geometries.
Modifies the geometry or metadata of specific features (points, lines) already on the map.
Deletes entire layers, individual elements, or even whole maps to keep your project clean and accurate.
Ask AI about this MCP
Supported MCP Clients
OAuth 2.0 Compatible Gemini Waiting for input…
Felt (Collaborative Maps): 11 Tools Available
These tools let you control every aspect of a map project—from listing existing maps to adding specific elements and updating entire data layers.
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 Felt (Collaborative Maps) on Vinkius019e3895add elements
Adds new geometric features like points and lines to a specified layer.
019e3895create layer
Uploads raw geographic data (CSV, KML, etc.) from a URL and builds it into a new map layer.
019e3895create map
Sets up an entirely new Felt map project with specific viewing boundaries.
019e3895delete element
Removes a single, identified geographic feature from the map.
019e3895delete layer
Deletes an entire layer of data from the map project.
019e3895delete map
Permanently deletes the entire collaborative map file.
019e3895get layer
Retrieves detailed information about an existing data layer on a specific map.
019e3895get map
Gets the full metadata and current state of a specific collaborative map file.
019e3895list maps
Shows a complete list of all maps you have access to in your account.
019e3895update element
Changes the coordinates or attributes of an existing feature on the map.
019e3895update layer
Modifies the properties, name, or style rules for a data layer.
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 Felt (Collaborative Maps), then connect any of our 5,000+ other servers whenever your AI needs more. One click, no limits.
- Use this MCP plus 5,000+ 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 Felt. 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
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 server provides 11 capabilities that interface natively with Claude, ChatGPT, Cursor, and any MCP client. No middleware. No custom integration required.
Manually tracking and updating project maps is a nightmare.
Today, if your team changes a boundary or adds a temporary marker, someone has to open the original source file, adjust the coordinates in Excel, save it as GeoJSON, then upload that new file into the mapping platform. This process involves at least five different tabs and three separate manual validation steps just to place one point.
With this MCP, you bypass all of that friction. You tell your agent what needs fixing—say, 'The service area boundary is off by 50 meters.' The agent uses its tools to find the correct layer and apply the necessary updates directly in conversation. It's immediate.
Felt (Collaborative Maps) MCP Gives You Full Control Over Your Geospatial Data
Gone are the days of waiting for a data engineer to process and upload new fields. You can now use `create_layer` with raw URLs, and your agent handles the schema definition and processing in the background.
What's different is that you own the workflow. The map becomes an active participant in your project management; it changes when you tell it to change.
What you can do with this MCP connector
Connect Felt to your AI client to take charge of collaborative mapping projects through natural conversation. Forget clicking through dozens of menus—you tell your agent what you need done with the map, and it handles the rest. You can ask it to list all existing maps, or even create a brand new one centered on specific coordinates.
Need to add data? Simply give instructions; your agent will handle uploading structured files like CSVs or GeoJSONs, building specialized layers in the process. If you find an element is wrong, you don't have to manually edit it; you just tell your agent to update a feature or delete an old boundary.
This capability makes complex spatial work possible directly from any compatible client connected through Vinkius.
This MCP lets you manage every detail of a map project—from setting up the initial viewports to programmatically adding and styling elements on existing layers.
019e3895-dfa5-71bc-b8e6-7803ed444b34 
How Felt (Collaborative Maps) MCP Works
- 1 First, subscribe to this MCP on Vinkius and provide the required Felt API Token.
- 2 Next, tell your agent what you need: whether it's listing existing maps or creating a new one with specific coordinates.
- 3 Finally, give clear instructions like 'Add these points to the 'Service Routes' layer,' and your agent executes the full data operation.
The bottom line is, you talk to your AI client, and it speaks directly to the map service to make changes on your behalf.
Who Is Felt (Collaborative Maps) MCP For?
This MCP is for professionals who spend time visualizing real-world data. You're the GIS Analyst stuck clicking through dozens of dashboards; you need a single interface where talking to an agent does the heavy lifting. It’s for anyone whose job involves taking raw coordinates and turning them into actionable visual maps.
Needs to quickly prototype map visualizations, upload various datasets (GeoJSON, CSV), and run spatial checks without writing boilerplate API calls.
Manages long-term project maps. They need to track changes—like adding new proposed infrastructure elements or updating service boundaries—as field data comes in.
Visualizes service areas and complex routes. They use this MCP to programmatically add temporary collection points or redraw operational zones on shared, live maps.
What Changes When You Connect
- Stop manually juggling data uploads. By using
create_layer, you simply give the agent a public URL, and it handles turning complex files like GeoJSON or KML into functional layers on the map. - Never lose context again. If your AI client needs to know what's currently on the screen before making a change—like listing all accessible maps with
list_maps—this MCP provides that immediate background detail. - Need to correct an old boundary? Instead of opening the source file, you just tell the agent to update it. Using
update_elementallows for precise fixes directly within your conversational workflow. - Keep projects clean. If a layer is obsolete or contains bad data, use
delete_layerordelete_mapto remove it instantly. This keeps your map workspace focused on what matters right now. - Streamline analysis context by using the getter tools. Calling
get_mapandget_layergives your agent all the necessary details—the metadata, the styling rules—to reason about spatial relationships correctly.
Real-World Use Cases
The Planner Needs to Check Scope
A city planner has five different project maps. Instead of logging into five separate dashboards, they ask their agent to 'list all my current Felt maps.' The agent replies with a list, letting the planner immediately jump to the right visualization and check the scope against local regulations.
The Scientist Needs New Data Visualized
A researcher finishes collecting survey data in a CSV file. They don't want to clean it up first. They tell their agent, 'Create a new layer using this URL.' The agent uses create_layer and uploads the data, making it visible on the map instantly for analysis.
The Ops Team Needs an Adjustment
A logistics team notices that one of their service boundaries is drawn incorrectly. Rather than manually editing the source file, they tell their agent to 'Update element X' at specific coordinates. The map boundary corrects itself immediately.
The Project Manager Needs a Clean Slate
A project concludes, leaving behind many temporary layers and old elements. The manager asks the agent to delete all non-core data using delete_layer and then uses get_map to confirm that only the final master layer remains.
The Tradeoffs
Trying to describe a map by text alone
Asking your agent, 'Tell me about the blue line on the bottom right.' The agent has no visual context and can't interpret vague directional language.
→
You must reference the layer name and element type. Say: 'Using get_layer details, find the elements in the 'Roads' layer that are marked as 'temporary'.'
Overwriting data without checking first
Telling your agent to 'Add this new element,' but forgetting if a similar element already exists. This results in conflicting or duplicate data on the map.
→
First, call get_map and list_maps to understand the current state. Then, use update_element instead of adding a new one if you intend to modify existing data.
Forgetting which map the action applies to
Simply asking to 'Delete that layer.' The agent won't know which project or map file you mean, leading to errors.
→
Always specify the scope. Begin by calling list_maps and naming the target map in every command.
When It Fits, When It Doesn't
Use this MCP if your primary workflow involves taking raw geographic data—like CSVs, KMLs, or GeoJSON files—and turning it into a structured, editable visualization on a collaborative map. The tools are perfect for the GIS analyst who needs to add elements (add_elements), manage layers (create_layer, delete_layer), and update coordinates without writing custom code.
Don't use this MCP if your goal is simply summarizing text documents or generating plain written reports. For pure data retrieval, you might need a general document analysis tool. If the problem is managing abstract relationships between different datasets (beyond spatial proximity), you should look for graph database tools instead of map manipulation tools.
Common Questions About Felt (Collaborative Maps) MCP
How do I list maps using the Felt (Collaborative Maps) MCP? +
You ask your agent to 'list all accessible maps.' The list_maps tool returns every map you have access to, letting you pick a starting point for work.
Can I add elements using the Felt (Collaborative Maps) MCP? +
Yes. You instruct your agent to use the add_elements tool and provide the coordinates and the target layer, so it places points or lines instantly.
What if I want to upload a big CSV file? +
You tell the agent to 'Create a new layer' and point it to the public URL for your CSV. The create_layer tool handles the data ingestion process for you.
How do I update an element on a map? +
You use the update_element tool, providing the ID of the feature and the new coordinates or attributes. This modifies the existing point without creating a duplicate.
Before using any map tool like `get_map` or `list_maps`, what credentials do I need to set up my agent? +
You must provide a Felt API Token during the initial setup. This token authorizes your AI client, giving it permission to read and modify your maps within the Felt workspace.
Using `update_layer`, how can I apply complex visual styling or change names? +
You update layer appearance by referencing the Felt Style Object (FSO) in your prompt. This allows you to programmatically define colors, visibility rules, and naming conventions for detailed data visualization.
If I want to archive an old project, is it safe to use `delete_map`? +
Yes, the delete_map tool handles complete removal of a map. However, this action permanently deletes all associated metadata and elements, so always confirm your intent first.
When I run `get_layer`, what specific details can my agent retrieve about the uploaded dataset? +
The tool returns detailed layer metadata, including the original data source type (GeoJSON, CSV), the current element count, and any applied styling rules. This context is crucial before making updates or additions.
Can I upload my own geographic data files to a map? +
Yes! Use the create_layer tool by providing a public URL to your GeoJSON, CSV, or KML file. The server will initiate the upload and processing into your specified Felt map.
How do I add a specific point or shape to an existing layer? +
You can use the add_elements tool. Simply provide the layer_id and a JSON array of GeoJSON features (points, lines, or polygons) you want to add to that layer.
Is it possible to change the visual style of a map layer? +
Yes. Use the update_layer tool and provide a 'Felt Style Object' (FSO) in the style parameter to programmatically change colors, icons, or visibility rules.
Use it with your favorite AI tools
Connect this server to Cursor, Claude, VS Code, and more.