An astronaut overseeing a futuristic data hub
Download PDFMarkdown Version

Connect Roboflow to ChatGPT, Claude and Gemini

Set it up once and it works in ChatGPT, Claude, Cursor, and any other client you love. It all runs safely in the background, so you can relax and focus on what really matters. Run your entire CV pipeline through your AI agent.

Ask AI about this Connector

Developed, maintained, and hosted by Vinkius.

MCP VERIFIED · PRODUCTION READY · VINKIUS GUARANTEED

Waiting for input…

Works with modern AI clients that support MCP, including ChatGPT, Claude, Cursor, and more.

ChatGPTClaudeCursorPerplexityGeminiMicrosoft CopilotRaycastMeta AI

Your AI does the work. We keep it safe and sound.

No other AI tool gives you this. Your workflow runs in the client you already love, your account stays protected, and you decide what the agent can and cannot do. It is the whole reason to make the switch.

Each Connector lives inside our infrastructure, the same place where your workflow runs. The three things below are already part of it, working quietly in the background so you never have to think about them. Nothing to add, nothing to set up, nothing to worry about.

A futuristic MCP client docking bay
V8 Isolate01

Each Connector runs inside its own sealed environment.

Each Connector runs inside its own sealed V8 runtime — no access to your files, your system, or other servers. A Connector never sees past its own request.

Kill Switch02

One switch stops every agent. Instantly.

One action deactivates every Connector, revokes every token, and ends every session at once. AI Governance keeps the record.

FinOps per User03

Spending limits that pause instead of surprise.

Every user keeps their own connections and credentials, isolated from everyone else. Anything that would cross your spending limit is held for your approval, not billed.

Three controls, built into the layer your AI connects through.

Curated set · 29 capabilities

We ship only the actions that matter.

Capability set01 / 08

01-04

4 capabilities in this set.

Part of 29 available through Roboflow.

  1. 01

    Search workspace images

    This capability lets you find and filter specific images across your entire workspace.

  2. 02

    Get dataset health

    Analyze your dataset to check for class distribution issues or missing annotations.

  3. 03

    Get image

    Retrieve detailed information about a single specific image.

  4. 04

    Get root

    Verify your connection and find your default workspace.

Capability set02 / 08

05-08

4 capabilities in this set.

Part of 29 available through Roboflow.

  1. 05

    Get training results

    Pull performance metrics like mAP and precision for a specific training version.

  2. 06

    Restore trash

    Bring an item back from the trash to your workspace.

  3. 07

    Run inference

    Execute a model on an image using hosted models to see real-time results.

  4. 08

    Search project images

    Filter and find specific images within a single project.

Capability set03 / 08

09-12

4 capabilities in this set.

Part of 29 available through Roboflow.

  1. 09

    Add projects to folder

    This capability moves specific projects into a designated folder within your workspace. It is available for Enterprise users.

  2. 10

    Auto label

    Start an automated labeling job using foundation models for a specific batch of images.

  3. 11

    Cancel training

    Stop a training job that is currently in progress.

  4. 12

    Get project

    Fetch metadata and version history for a specific project.

Capability set04 / 08

13-16

4 capabilities in this set.

Part of 29 available through Roboflow.

  1. 13

    Create annotation job

    Assign a batch of images to a specific labeler and reviewer to get your data annotated.

  2. 14

    Create folder

    Create a new project folder in your workspace. This is an Enterprise feature.

  3. 15

    Create project

    Set up a brand new project within a specified workspace.

  4. 16

    Delete images

    Remove multiple images from a project at once using their IDs.

Capability set05 / 08

17-20

4 capabilities in this set.

Part of 29 available through Roboflow.

  1. 17

    Delete project

    Move a project or a specific version to the trash.

  2. 18

    Download dataset

    Get a download link for a zipped version of your dataset in formats like YOLO or COCO.

  3. 19

    Fork universe project

    Copy a public project from Roboflow Universe into your own workspace.

  4. 20

    Get async task

    Check the status of long-running operations like large exports or project forks.

Capability set06 / 08

21-23

3 capabilities in this set.

Part of 29 available through Roboflow.

  1. 21

    Get version

    Get the metadata associated with a specific dataset version.

  2. 22

    List folders

    View all project folders in your workspace. This is an Enterprise feature.

  3. 23

    List trash

    See all the items you have recently deleted in your workspace.

Capability set07 / 08

24-26

3 capabilities in this set.

Part of 29 available through Roboflow.

  1. 24

    List workspace projects

    Use this to get a full list of projects and their details for a specific workspace.

  2. 25

    Manage image tags

    Add or remove tags from specific images to help organize your data.

  3. 26

    Start training

    Trigger a new model training run on a specific dataset version.

Capability set08 / 08

27-29

3 capabilities in this set.

Part of 29 available through Roboflow.

  1. 27

    Stop training

    Early stop a training job that is currently running.

  2. 28

    Upload annotation

    Attach an annotation file to an image that is already in your project.

  3. 29

    Upload image

    Add new image data to a project using a batch name.

Real work

What teams actually run

01

Rapid Prototyping

Fork a project from Roboflow Universe and start training immediately via your agent.

02

Data Quality Audits

Use the agent to check class distribution and find missing annotations in a dataset.

03

Continuous Training

Automate the process of uploading new images and starting new training runs.

04

Model Validation

Run inference on new images to verify that a trained model meets requirements.

Automate a task

Put Roboflow to work on one thing.

Step-by-step automations your AI runs through this Connector: the workflow, the prompts to start with, and the exact capabilities behind each move.

One connector, every AI

Roboflow works with the most popular AI clients.

These are the most popular clients, each with a step-by-step guide: one link, set up once, with governance and visibility built in. And because everything runs on the MCP standard, the same connection also works in any other compatible client — nothing to rebuild.

ClaudeClaudeChatGPTChatGPTGeminiGeminiPerplexityPerplexityGrokGrokMicrosoft CopilotMicrosoft CopilotCursorCursorVS CodeVS CodeWindsurfWindsurfJetBrainsJetBrainsClineClineLangChainLangChainVercel AI SDKVercel AI SDKLovableLovableZ.aiZ.aiRaycastRaycastQwen CodeQwen CodeKimi CodeKimi CodeLe ChatLe Chat

Building your own app? The connector is yours to use.

You don't need a client to put Roboflow to work: the same hosted connection plugs into your own applications and agent code, with the same governance on every request. Build with it, chat with it — one connection for both.

Observed, not estimated

1184ms average. Fast in production.

Roboflow is checked daily against the live service.

Daily averagePeak 1512ms
Oct 3Today
Fastest day
915ms
Slowest day
1512ms
14-day trend
Improving-9%

Connect your client

One URL. Every client.

Activate the Connector, copy your link, and paste it into the client you already use. 29 capabilities arrive ready to run.

Preview access · not provider authentication

The vk_preview_* token belongs to Vinkius preview infrastructure. It lets Claude discover and display the capabilities of Roboflow, so you can see the experience inside your AI.

It does not authenticate your account with Roboflow. Actions requiring credentials or live account data may not run until you activate the Connector and authorize the service.

Roboflow Connector

You're all set. Choose your MCP client and follow the setup instructions.

Connector linkhttps://edge.vinkius.com/vk_preview_3PcAB3lxgco6m9RlryrdB51L1KU7rDlnqGHOwOaP/mcp

Claude Desktop

Follow the steps below to connect in seconds.

  1. 1In Claude Desktop, open Settings → Connectors.
  2. 2Click “Add custom connector” and paste the connector link above as the remote MCP server URL.
  3. 3Click Add and start a new chat — Roboflow capabilities are ready to use.
Configuration · claude_desktop_config.jsonCopy
{
  "mcpServers": {
    "roboflow-mcp": {
      "url": "https://edge.vinkius.com/vk_preview_3PcAB3lxgco6m9RlryrdB51L1KU7rDlnqGHOwOaP/mcp"
    }
  }
}
  • Claude
  • ChatGPT
  • Cursor
  • VS Code
  • Windsurf
  • Claude Code
  • JetBrains
  • Cline

Step-by-step instructions for each client are in the guide. How to connect

Guided setup for Claude? Set up Roboflow in Claude and other AIs

See all the AI clients this connector works with ↑

Who it's for

Built for the work Roboflow owners hand off.

This MCP is built for technical professionals managing computer vision lifecycles.

  • 01

    ML Engineers

    Monitor training progress and dataset health without leaving the IDE.

  • 02

    Data Scientists

    Query dataset versions and export data for custom training scripts.

  • 03

    Product Teams

    Audit model performance and check inference results through conversation.

FAQ

Questions Roboflow owners ask.

  • 01

    How do I connect my Roboflow account?

    You just need to provide your Roboflow Private API Key after subscribing to the MCP on Vinkius.

  • 02

    Which AI clients can I use with this MCP?

    You can use any MCP-compatible client, including Claude, Cursor, and Windsurf.

  • 03

    Can I run model training through my agent?

    Yes, you can start, stop, or cancel training runs for specific project versions.

  • 04

    Does this support Enterprise features?

    Yes, capabilities like creating folders and adding projects to folders are available for Enterprise users.

  • 05

    Can I download my datasets?

    Yes, the agent can generate download links for zipped datasets in formats like YOLO or COCO.

  • 06

    How can I verify if my Roboflow API key is correctly configured?

    You can use the get_root capability. It will attempt to authenticate with your key and return the default workspace details if successful.

  • 07

    Can I get the training performance metrics for a specific model version?

    Yes! Use the get_training_results capability by providing the workspace, project, and version ID. It returns mAP, precision, recall, and other training metrics.

  • 08

    Is it possible to export my dataset to a specific format like YOLOv5?

    Absolutely. Use the download_dataset capability and specify the format parameter (e.g., 'yolov5pytorch') to receive a download link for your zipped dataset.

Explore

More in Developer Tools

Suggestions

Connect Roboflow to Claude, Cursor, ChatGPT & more