How to Use the DVC MCP in Pydantic AI

Q: Can I use Pydantic AI to filter DVC views?

Yes. The agent can call listviews to get all slices, validate the list, and then request details for a specific slice using getview. Every step is strictly typed.

Validate every single DVC experiment and dataset view at runtime using type-safe Pydantic AI.

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 Pydantic AI

Create your Vinkius account to connect DVC to Pydantic AI 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 Pydantic AI

Ask AI about this MCP

ChatGPT

Claude

Perplexity

Enforce strict schemas on DVC MCP Server runs

The `list_experiments` tool returns structured DVC run data that Pydantic AI validates against strict Python types at runtime. If a DVC run payload contains unexpected fields, Pydantic AI catches it immediately. You don't get silent failures in your downstream evaluation scripts. This strictness is crucial when parsing complex DVC metrics in Pydantic AI. Your agent can confidently extract model accuracy numbers knowing the DVC data structure matches your expected schema.

Type-check dataset views before training

Use `get_view` and `list_views` to verify your DVC data splits before starting a model run in Pydantic AI. The Pydantic AI agent validates the DVC view structure using Pydantic models to ensure required features are present. If a view's missing a required column, this MCP tool execution raises a Pydantic AI validation error. The Pydantic AI agent can then automatically flag the issue to the user instead of proceeding with broken training.

Verify project and user profiles

The `get_project` and `get_user` tools fetch metadata about your DVC ML workspace into Pydantic AI. Pydantic AI parses these payloads into clean Python objects, making it easy to check DVC project status or verify the run owner's identity. Combine this with `list_projects` to build a type-safe dashboard of your active DVC repositories. Your Pydantic AI agent can monitor workspace health without dealing with raw, unvalidated JSON.

Setup guide

Set up DVC MCP in Pydantic AI

Prerequisites

Python 3.10+ installed
pydantic-ai-slim[fastmcp] package
Active Vinkius subscription with a valid endpoint token

1

Install Pydantic AI with FastMCP
Run pip install "pydantic-ai-slim[fastmcp]". The FastMCP toolset replaces the deprecated MCPServerHTTP class with full protocol support.
2

Configure the FastMCPToolset
Pass a JSON-style config dict to FastMCPToolset with your Vinkius URL. Replace [YOUR_TOKEN_HERE] with your token from cloud.vinkius.com. Supports Streamable HTTP, SSE, and Stdio transports.
3

Create and run your agent
Pass the toolset to Agent(toolsets=[toolset]) and call agent.run(). Swap openai:gpt-4o for any supported model — Anthropic, Google, Mistral, or Groq.

agent.py

from pydantic_ai import Agent
from pydantic_ai.toolsets.fastmcp import FastMCPToolset

toolset = FastMCPToolset({
    "mcpServers": {
        "dvc-mcp": {
            "url": "https://edge.vinkius.com/[YOUR_TOKEN_HERE]/mcp"
        }
    }
})

agent = Agent(
    "openai:gpt-4o",
    toolsets=[toolset],
    system_prompt="You have access to DVC tools.",
)

result = await agent.run("List recent DVC transactions")
print(result.output)

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 Pydantic AI

Use the unified MCPToolset class pointing to your Vinkius HTTP endpoint. Pass this toolset to your Agent constructor. The agent can then call `list_projects` and other tools with full runtime validation.

Pydantic AI will raise a validation error instantly. Instead of your agent hallucinating or acting on corrupt data from `list_experiments`, the execution halts loudly so you can fix the schema mismatch.

Yes. The agent can call `list_views` to get all slices, validate the list, and then request details for a specific slice using `get_view`. Every step is strictly typed.

No. The MCP server runs in a remote Vinkius sandbox. Your Pydantic AI agent communicates with it over HTTP, meaning you do not need local DVC CLI dependencies in your production environment.

We run the MCP connection inside an ephemeral V8 sandbox. Your project schemas, experiment metrics, and user profiles are processed in memory and never written to persistent disk.

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