How to Use the DVC MCP in Cursor

Q: Does Cursor need write access to my DVC repository?

No, these tools are read-only. Cursor uses getproject and listexperiments solely to read metadata and guide your code generation.

Q: How does Cursor handle massive DVC project registries?

The listprojects tool returns a clean list of active workspaces to keep the editor fast. Cursor only pulls detailed metrics when you ask it to analyze a specific project.

Inject live DVC experiment metrics directly into your Cursor agent loops.

See Vinkius in Action

Works with every AI agent you already use

…and any MCP-compatible client

MCP Servers - Free for Subscribers

Connect DVC MCP to Cursor

Create your Vinkius account to connect DVC to Cursor and route execution through our secure gateway. The platform manages server hosting, runtime updates, and security layers. Configuration requires no manual server provisioning.

GDPR Free for Subscribers

Setup DVC with Cursor

Ask AI about this MCP

ChatGPT

Claude

Perplexity

Inspect workspace states inside Cursor

The `get_project` tool provides the Cursor agent with the exact state of your active DVC directory. Your agent reads this structure to write training scripts that match your actual data paths. Instead of writing code against outdated folder structures, Cursor uses this live context. Your generated Python scripts run correctly on the first try.

Compare training runs with the DVC MCP Server

The `list_experiments` tool feeds your model metrics and hyperparameter logs directly to the Cursor composer. Your agent reads these runs to suggest code changes for your next training iteration. This eliminates manual comparison work. Cursor analyzes the differences between your best run and your current code, then modifies the training loop.

Map data splits during active coding

The `get_view` and `list_views` tools expose your data registry structures directly to your Cursor editor. Your agent inspects these views to verify that your data loader code targets the correct directories. This prevents path errors before you hit run. The MCP integration writes the exact loading logic based on the actual view definitions stored in your registry.

Setup guide

Set up DVC MCP in Cursor

Prerequisites

Cursor installed (macOS, Windows, or Linux)
Active Vinkius subscription with a valid endpoint token

1

Open MCP Settings

Go to Cursor Settings → MCP or open the Command Palette (Cmd+Shift+P / Ctrl+Shift+P) and search for "MCP: Add Server".
2

Add the DVC MCP

Cursor will create or open .cursor/mcp.json in your project root. Paste the JSON snippet on the right. Replace [YOUR_TOKEN_HERE] with your endpoint token from cloud.vinkius.com.
3

Enable Agent mode

Open Composer (Cmd+I / Ctrl+I) and switch to Agent mode using the dropdown at the top. MCP tools are only available in Agent mode.
4

Verify the connection

Ask Cursor something like "List my recent DVC transactions." If the MCP tools are loaded correctly, Cursor will call the DVC tools automatically. You can also check Settings → MCP for a green status indicator.

.cursor/mcp.json

{
  "mcpServers": {
    "dvc-mcp": {
      "url": "https://edge.vinkius.com/[YOUR_TOKEN_HERE]/mcp"
    }
  }
}

Get your connection token →

Independent Platform Disclaimer: Vinkius is an independent platform and is not affiliated with, endorsed by, sponsored by, verified by, or otherwise authorized by DVC. 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.

Why Choose Vinkius

Vinkius connects your tools to AI with real-time monitoring and automatic cost savings — all from one dashboard.

Connect DVC now

Real-time monitoring

Live

visibility into every interaction

Connect your favorite tools to your AI and see exactly what's happening — every request, every response, in real time.

Built-in savings

60%

lower AI costs

Vinkius compresses data between your apps and your AI automatically. Lower bills every month — no configuration required.

Single dashboard

One

place for every integration

Every tool your AI connects to, managed from a single screen. One account, complete control.

Common questions about DVC MCP in Cursor

Add the server configuration to your global or project-level config file in Cursor. Once saved, this MCP toolset is active.

No, these tools are read-only. Cursor uses `get_project` and `list_experiments` solely to read metadata and guide your code generation.

No, this server does not write commits. It reads existing runs so Cursor can help you write the code for your next experiment.

The `list_projects` tool returns a clean list of active workspaces to keep the editor fast. Cursor only pulls detailed metrics when you ask it to analyze a specific project.

Yes, your metrics and pipeline definitions are never uploaded to external servers. The Vinkius MCP sandbox ensures your project structure remains completely private.

Use it with your favorite AI tools

Connect this server to Cursor, Claude, VS Code, and more.

OpenAI Agents SDK sdk-python

Google ADK sdk-python

Pydantic AI sdk-python

Vercel AI SDK sdk-typescript