How to connect Roboflow to Cursor

  • Cursor gets 21 Roboflow tools through one Metorial MCP URL.
  • Setup is 3 steps; teammates then sign in with SSO.
  • You need Cursor with MCP support.

What Cursor can do with Roboflow

Ask your agent in plain language. Each of these maps to a real tool on the server.

  • Ask Cursor to use the create project tool in Roboflow
  • Ask Cursor to use the create version tool in Roboflow
  • Ask Cursor to use the search images tool in Roboflow
  • Ask Cursor to use the upload image tool in Roboflow
  • Ask Cursor to use the get version tool in Roboflow

Add the Roboflow MCP server to Cursor

Run these once. Afterwards, everyone you give access to signs in the same way.

  1. 1. Enable Roboflow in Metorial

    In your Metorial portal, enable the Roboflow MCP server and sign in with the account that owns the data. Access stays governed by your team policies.

  2. 2. Add Metorial to your MCP config

    Use ~/.cursor/mcp.json for every project, or .cursor/mcp.json for one.

    {
    "mcpServers": {
    "metorial": {
    "url": "<your-magic-mcp-url>"
    }
    }
    }
  3. 3. Sign in

    Open Customize in the sidebar, find Metorial and sign in to load Roboflow tools.

Give your team Roboflow in Cursor

Once Roboflow is enabled in Metorial, teammates sign in with SSO and get the same connection in Cursor, with only the tools their role allows.

One link covers every app you approve. See how the MCP gateway handles access, roles and audit logs for the whole company.

Roboflow tools available in Cursor

Full descriptions and inputs are on the Roboflow MCP server page.

  • List Annotation Jobs: List all annotation jobs for a project.
  • Upload Annotation: Upload an annotation file for a specific image in a project.
  • Run Inference: Run predictions on an image using a trained model deployed on the Roboflow hosted inference API.
  • Create Annotation Job: Create a new annotation job for a project.
  • Get Project: Get detailed information about a specific Roboflow project, including its dataset versions, annotation classes, image splits, and training status.
  • Train Model: Start training a model on a specific dataset version using Roboflow Train.
  • List Projects: List all projects in a Roboflow workspace.
  • Export Dataset: Export a dataset version as a downloadable ZIP in a specified annotation format.
  • Get Image: Get detailed information about a specific image in a project, including its annotations, labels, tags, split assignment, and URLs for the original and thumbnail versions.
  • Delete Images: Delete one or more images from a Roboflow project.
  • Create Project: Create a new computer vision project in the workspace.
  • Create Version: Generate a new versioned snapshot of a project's dataset with optional preprocessing and augmentation settings.
  • Search Images: Search for images within a project using text prompts, visual similarity to an existing image, or by filtering on tags, classes, and batch assignments.
  • Upload Image: Upload an image to a Roboflow project.
  • Get Version: Get detailed information about a specific dataset version, including preprocessing and augmentation settings, training results (mAP, precision, recall), and model endpoint details.
  • And 6 more on the Roboflow MCP server page.

Using Roboflow tools in Cursor

What to expect once Roboflow is connected.

  • Cursor picks MCP tools automatically when they are relevant, including in Plan Mode. Name a Roboflow tool to steer it.
  • Cursor asks for approval before it uses an MCP tool by default. Click the arrow next to the tool name to check the arguments first.
  • Switch the Metorial server off from Customize in the sidebar without removing it.
  • Cursor shows each Roboflow tool response in chat, with expandable views of the arguments and the result.

Troubleshooting Cursor

Metorial connects but tools are missing or failing
Open the Output panel (Cmd+Shift+U), choose MCP Logs, and look for authentication or connection errors.
Your company blocks the server
Enterprise admins approve remote servers in Team Settings, then MCP Configuration. Ask them to add a URL entry for your Metorial URL.
Cursor asks before every tool call
That is the default. Allowlist the Metorial tools you trust in your Run Mode settings to run them without a prompt.

Based on the Cursor MCP documentation.

Frequently asked questions

Common questions about connecting Roboflow to AI agents with Metorial.

  1. Does Cursor support the Roboflow MCP server?
    Yes. Cursor connects to the Roboflow MCP server through its MCP config file, pointed at your Metorial URL. You need Cursor with MCP support.
  2. No. You sign in to Roboflow once through Metorial, and Cursor uses that connection. No API keys go into Cursor's settings.
  3. All 21 Roboflow tools your team allows, such as List Annotation Jobs, Upload Annotation, Run Inference. Admins can narrow that list per team.
  4. Yes. Role-based access control decides who can use Roboflow and which of its tools appear in Cursor, and every call is logged with the person's identity.
  5. Open the Output panel (Cmd+Shift+U), choose MCP Logs, and look for authentication or connection errors.