How to Use the H2O.ai MCP in Mastra AI

Q: How do I integrate H2O.ai with Mastra AI?

Install @mastra/mcp and instantiate a new MCPClient pointing to your Vinkius endpoint. You can then register the server tools directly with your Mastra agents. The framework automatically handles transport negotiation using SSE or HTTP.

Q: Can a Mastra AI agent automatically restart failed H2O.ai jobs?

Yes, you can design a workflow that monitors jobs using listjobs. If a job transitions to a failed state, the agent can analyze the logs and use its toolset to handle the escalation or retry sequence.

Q: How does Mastra AI handle human-in-the-loop for H2O.ai tasks?

You can enable requireToolApproval on critical tools like getmodel or cluster inspection actions. This ensures that your agent pauses and requests developer confirmation before proceeding with expensive operations.

Build resilient TypeScript workflows that monitor H2O.ai jobs and handle cluster errors using Mastra AI and this MCP Server.

See Vinkius in Action

Works with every AI agent you already use

…and any MCP-compatible client

MCP Servers - Free for Subscribers

Connect H2O.ai MCP to Mastra AI

Create your Vinkius account to connect H2O.ai to Mastra 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 H2O.ai with Mastra AI

Ask AI about this MCP

ChatGPT

Claude

Perplexity

Build resilient ML pipelines with Mastra AI

The `list_jobs` tool lets your TypeScript agents run inside structured Mastra workflows to track model training from start to finish via MCP. Stop writing fragile shell scripts to monitor your training cycles. If a job fails, Mastra handles the retry logic automatically. The agent can examine the failure state, check cluster health, and determine whether to trigger a restart or alert the engineering team.

Automated cluster resource guardrails

The `cloud_status` tool verifies available memory and active node counts before your agent kicks off a heavy training pipeline. This programmatic check keeps your cloud costs predictable and prevents queue bottlenecks. If the cluster is overloaded, the workflow engine pauses execution or diverts the job. You avoid system crashes by letting the agent make routing decisions in real-time.

Validate training data frames automatically

The `get_frame` tool inspects the schema and row counts of your datasets before starting any training run. Your agent uses `list_frames` to locate the correct file and checks for missing columns. If the frame contains missing columns or incorrect types, the agent halts the workflow. This prevents wasted compute hours on models that are bound to fail.

Setup guide

Set up H2O.ai MCP in Mastra AI

Prerequisites

Node.js 18+ and a TypeScript project
@mastra/mcp + @mastra/core packages
Active Vinkius subscription with a valid endpoint token

1

Install dependencies
Run npm install @mastra/mcp @mastra/core plus your preferred model provider (e.g. @ai-sdk/openai).
2

Configure the MCPClient
Create an MCPClient with your Vinkius endpoint as a URL object. Replace [YOUR_TOKEN_HERE] with your token from cloud.vinkius.com.
3

Discover and inject tools
Call mcpClient.listTools() and spread the result into your agent's tools object. All H2O.ai tools become native Mastra tools.
4

Run with any model
Swap openai("gpt-4o") for any AI SDK-compatible provider. Call agent.generate() and the agent routes tool calls through MCP automatically.

agent.ts

import { MCPClient } from "@mastra/mcp";
import { Agent } from "@mastra/core/agent";
import { openai } from "@ai-sdk/openai";

const mcpClient = new MCPClient({
  id: "h2oai-mcp-client",
  servers: {
    "h2oai-mcp": {
      url: new URL(
        "https://edge.vinkius.com/[YOUR_TOKEN_HERE]/mcp"
      ),
    },
  },
});

const agent = new Agent({
  name: "H2O.ai Agent",
  model: openai("gpt-4o"),
  instructions: "You have access to H2O.ai tools.",
  tools: {
    ...(await mcpClient.listTools()),
  },
});

const result = await agent.generate(
  "List recent H2O.ai transactions"
);
console.log(result.text);

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 H2O.ai. 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 H2O.ai 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 H2O.ai MCP in Mastra AI

Install `@mastra/mcp` and instantiate a new `MCPClient` pointing to your Vinkius endpoint. You can then register the server tools directly with your Mastra agents. The framework automatically handles transport negotiation using SSE or HTTP.

Yes, you can design a workflow that monitors jobs using `list_jobs`. If a job transitions to a failed state, the agent can analyze the logs and use its toolset to handle the escalation or retry sequence.

You can enable `requireToolApproval` on critical tools like `get_model` or cluster inspection actions. This ensures that your agent pauses and requests developer confirmation before proceeding with expensive operations.

Absolutely. You can chain actions together, such as fetching data schemas with `get_frame` and comparing them against target definitions. The agent makes decisions at each step based on the tool outputs.

Yes, your model configurations and training parameters retrieved via `get_model` are protected. All tool executions occur within a zero-trust, ephemeral sandbox. No sensitive metadata is logged or cached on external servers.

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