Roboflow + Claude Code

Manage datasets, train models, build Workflows and run inference in Roboflow from Claude Code, with the Roboflow plugin and MCP server.

Claude Code is Anthropic's coding agent for the terminal. The Roboflow plugin adds Roboflow's hosted MCP server and Skills, task guides the agent reads as it works, so Claude Code can use your Roboflow workspace while it codes.

Install the plugin from the terminal, sign in to Roboflow, and ask in plain language. Claude Code can work with your datasets and annotations, train and evaluate models, build Workflows, run inference, and write the code that calls your model from your app.

How to connect Claude Code to Roboflow

5 steps
  1. Add the Roboflow marketplace

    In your terminal, run claude plugin marketplace add roboflow/computer-vision-skills.

  2. Install the Roboflow plugin

    Run claude plugin install roboflow. This adds the Roboflow MCP server and Skills.

  3. Sign in to Roboflow

    The first time Claude Code connects, Roboflow opens in your browser so you can sign in and authorize it.

  4. Check the connection

    Start Claude Code and ask it to list the projects in your Roboflow workspace.

  5. Ask Claude Code to build

    Ask it to run a model on your images, train a model, build a Workflow, or add inference code to your app.

What teams build with Claude Code

  • Add vision to an app you're building

    Have Claude Code write the code that calls your Roboflow model, right in your repo.

  • Train a model from the terminal

    Manage your dataset and versions, then train a model and check its evaluation results.

  • Build a Workflow from a description

    Describe the pipeline and let Claude Code build and save the Workflow.

Questions

No. Claude Code connects with OAuth and signs in with your Roboflow account in the browser.

Yes. Run claude plugin install roboflow --scope local to add it to the current project only.

No. The MCP server and Skills have no extra cost. You pay only for the Roboflow usage behind them on your plan.

Why are companies choosing Roboflow?

The end-to-end platform for vision AI: create datasets, train models and deploy to production.

  • Flexible deployment modes

    Move from a scalable cloud API to edge hardware like NVIDIA Jetson without managing complex dependencies.

  • Integrated workflow builder

    Combine multiple AI models, custom business logic and third-party integrations into one production-ready pipeline in a low-code editor.

  • Model monitoring and alerts

    Dashboards track confidence, inference time and request volume for your deployed models, and email alerts tell your team when confidence suddenly drops.

View all
Get started

Build on the Platform

For developers, engineers, and technical founders who want to get hands on. Try the free tier; the docs are open.

Bring it into your operation

For heads of AI, operations leaders, and enterprise teams. Bring a known problem, or work with us to find the first one worth taking on.