# Roboflow MCP server

> Manage Roboflow computer vision datasets, annotation, workflows, training and inference from your coding agent.

- Listing: https://mcp.tc/i/roboflow
- Connect: use the server's own URL `https://mcp.roboflow.com/mcp` (OAuth sign-in at the server); clients connect to it directly. The listing link is a page, not an MCP endpoint.
- Type: remote (Streamable HTTP)
- Auth: OAuth sign-in
- Category: [AI & Machine Learning](https://mcp.tc/c/ai-ml)
- Vendor: Roboflow
- Homepage: <https://roboflow.com/mcp>

## About

Connects an assistant to a Roboflow workspace. It can manage projects and images, save annotations, run auto-labeling jobs, work with Workflows, launch project deployments, run inference and chat with the Roboflow AI agent. The vendor site lists 163 tools grouped by task, from dataset to deployment.

Runs as a hosted streamable HTTP endpoint at https://mcp.roboflow.com/mcp. Users sign in to Roboflow with OAuth. A Roboflow API key sent in an x-api-key header also works as an alternative.

## What it can do

- Create, list, fork and manage projects in a workspace
- Upload, search and update metadata for images
- Save annotations and run hosted auto-labeling jobs
- Manage annotation batches, assignments and review
- Launch project deployments and run inference
- Analyze images and short videos to plan CV tasks
- Chat with the Roboflow AI agent to build Workflows

## Example prompts

- "List the projects in my Roboflow workspace"
- "Upload these images to my defect-detection project"
- "Start an auto-label job on the latest batch of images"
- "Launch a deployment for my project and run inference on a test image"

## Install

### Claude Code

1. Run this in a terminal, in your project folder:

```bash
claude mcp add --transport http roboflow https://mcp.roboflow.com/mcp
```

2. Start Claude Code, type `/mcp`, pick **roboflow** and choose **Authenticate**. A browser window opens for the Roboflow sign-in.

Add `--scope user` to make it available in every project, not just this one.

### Claude Desktop

1. Open **Settings → Connectors** and click **Add custom connector**.

2. Name it **Roboflow** and paste this URL:

```url
https://mcp.roboflow.com/mcp
```

3. Click **Add**, then **Connect**, and sign in when Roboflow asks.

Claude Desktop’s JSON config file only starts local servers. Remote servers go through Connectors, and connectors you add on claude.ai show up here too.

### claude.ai

1. Open the connector form on claude.ai. This button fills in the name and URL for you:

[Add to claude.ai](<https://claude.ai/customize/connectors?modal=add-custom-connector&connectorName=Roboflow&connectorUrl=https%3A%2F%2Fmcp.roboflow.com%2Fmcp>) (opens connector settings)

2. Check that the URL reads `https://mcp.roboflow.com/mcp` and click **Add**.

3. Click **Connect** and sign in when Roboflow asks.

Free plans allow one custom connector. On Team and Enterprise plans an owner adds it under **Organization settings → Connectors**.

### ChatGPT

1. On chatgpt.com, open **Settings → Security and login** and turn on **Developer mode**.

2. Go to `chatgpt.com/plugins` and click **+** to create an app for a remote MCP server.

3. Paste `https://mcp.roboflow.com/mcp` as the server URL and choose **OAuth**. ChatGPT sends you to Roboflow to sign in.

Developer mode is available on the web for Plus, Pro, Business, Enterprise and Education accounts.

### Cursor

[Add to Cursor](<https://cursor.com/install-mcp?name=roboflow&config=eyJ1cmwiOiJodHRwczovL21jcC5yb2JvZmxvdy5jb20vbWNwIn0%3D>) (opens Cursor)

Or add it by hand to `~/.cursor/mcp.json` (all projects) or `.cursor/mcp.json` (this project):

`mcp.json`:

```json
{
  "mcpServers": {
    "roboflow": {
      "url": "https://mcp.roboflow.com/mcp"
    }
  }
}
```

Cursor shows **Needs login** next to the server. Click it to sign in.

### VS Code

[Install in VS Code](<https://vscode.dev/redirect/mcp/install?name=roboflow&config=%7B%22type%22%3A%22http%22%2C%22url%22%3A%22https%3A%2F%2Fmcp.roboflow.com%2Fmcp%22%7D>) (opens VS Code)

Or from a terminal:

```bash
code --add-mcp '{"name":"roboflow","type":"http","url":"https://mcp.roboflow.com/mcp"}'
```

Or commit it to the repo in `.vscode/mcp.json`:

`.vscode/mcp.json`:

```json
{
  "servers": {
    "roboflow": {
      "type": "http",
      "url": "https://mcp.roboflow.com/mcp"
    }
  }
}
```

VS Code asks you to sign in the first time the server starts.

### Devin Desktop

1. Add it to `~/.config/devin/mcp_config.json` (macOS and Linux) or `%APPDATA%\devin\mcp_config.json` (Windows):

`mcp_config.json`:

```json
{
  "mcpServers": {
    "roboflow": {
      "serverUrl": "https://mcp.roboflow.com/mcp"
    }
  }
}
```

2. Refresh the MCP server list in Cascade and sign in when asked.

Devin Desktop is the new name for Windsurf. It reads `serverUrl` (or `url`) for remote servers.

### Codex

```bash
codex mcp add roboflow --url https://mcp.roboflow.com/mcp
codex mcp login roboflow
```

Or edit `~/.codex/config.toml` directly:

`config.toml`:

```toml
[mcp_servers.roboflow]
url = "https://mcp.roboflow.com/mcp"
```

### Gemini CLI

```bash
gemini mcp add --transport http roboflow https://mcp.roboflow.com/mcp
```

Then, inside Gemini CLI, run `/mcp auth roboflow` to sign in.

This adds it to the current project. Add `-s user` to use it everywhere.

### Any client

Most clients accept this shape. Some name the URL field differently: `serverUrl` in Devin Desktop, `httpUrl` in Gemini CLI’s settings file.

```json
{
  "mcpServers": {
    "roboflow": {
      "type": "http",
      "url": "https://mcp.roboflow.com/mcp"
    }
  }
}
```

Zed puts servers under `context_servers` in its settings. Cline needs `"type": "streamableHttp"`, or it assumes SSE.

Client only starts local servers? Bridge it with `npx -y mcp-remote https://mcp.roboflow.com/mcp`.

## Details

- Server version: 1.0.3
- Last checked: 2026-10-03 (reachable, asks for credentials)
- Listed: 2026-10-03
- Updated: 2026-10-03

---
Source: https://mcp.tc/i/roboflow (mcp.tc is an independent directory, not affiliated with this server's publisher). Corrections: https://mcp.tc/report