How to Use the Comet ML MCP in LangChain

Q: How do I connect the Comet ML MCP Server to my LangChain agent?

Install the langchain-mcp-adapters package and initialize MultiServerMCPClient with the server URL. Once connected, call client.gettools() and pass them directly to your agent constructor to start tracking runs.

Let your LangChain agents inspect model metrics and trace run parameters directly inside your chains using the Comet ML 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 Comet ML MCP to LangChain

Create your Vinkius account to connect Comet ML to LangChain 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 Comet ML with LangChain

Ask AI about this MCP

ChatGPT

Claude

Perplexity

Chain Comet ML workspace lookups in LangChain

This tool lets your LangChain agents call `list_workspaces` to inspect your machine learning workspaces without writing glue code. The run-loop locates the correct environment and immediately passes that context to `list_projects` in the next step of your chain. By linking these steps, your agent gets a clean path to find where your models are training. LangSmith traces the entire sequence, showing you exactly how the agent navigated your Comet workspaces to find the right project.

Debug training runs with active LangChain tool tracing

Your agent uses `get_experiment_params` to pull down the exact hyperparameters of a training run directly inside your chain. Stop guessing why a training run diverged and let your agent pass those parameters to your evaluation chains to see what went wrong. Because this runs over an MCP Server connection, LangChain manages the tool schemas out of the box. You get clean JSON payloads containing the precise API taxonomy types, which your agent can immediately parse and act upon.

Automate metric checks inside LangChain pipelines

This tool lets your agent call `get_experiment_metrics` to grab numeric bounds from a run and monitor performance automatically. If those metrics fall outside your target range, the agent calls `get_experiment` to fetch the full payload. This gives your agent the raw data it needs to decide whether to register a model or flag it for human review. It turns static training logs into active inputs for your decision-making chains.

Setup guide

Set up Comet ML MCP in LangChain

Prerequisites

Python 3.10+ installed
langchain-mcp-adapters + langgraph packages
Active Vinkius subscription with a valid endpoint token

1

Install dependencies
Run pip install langchain-mcp-adapters langgraph langchain-openai. The MCP adapters package converts MCP tools into native LangChain BaseTool objects.
2

Connect via HTTP transport
Use MultiServerMCPClient with "transport": "http" pointing to your Vinkius endpoint. Replace [YOUR_TOKEN_HERE] with your token from cloud.vinkius.com.
3

Create a ReAct agent
Pass the discovered tools to create_react_agent() from LangGraph. The agent automatically routes Comet ML tool calls through the MCP protocol.
4

Run with any LLM
Swap ChatOpenAI for ChatAnthropic, ChatGoogleGenerativeAI, or any LangChain-compatible model. The MCP tools work identically across all providers.

agent.py

from langchain_mcp_adapters.client import MultiServerMCPClient
from langgraph.prebuilt import create_react_agent
from langchain_openai import ChatOpenAI

async with MultiServerMCPClient({
    "comet-ml-mcp": {
        "transport": "http",
        "url": "https://edge.vinkius.com/[YOUR_TOKEN_HERE]/mcp",
    }
}) as client:
    tools = client.get_tools()

    agent = create_react_agent(
        ChatOpenAI(model="gpt-4o"),
        tools,
    )
    result = await agent.ainvoke({
        "messages": "List recent Comet ML transactions"
    })
    print(result["messages"][-1].content)

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 Comet ML. 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 Comet ML 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 Comet ML MCP in LangChain

Install the `langchain-mcp-adapters` package and initialize `MultiServerMCPClient` with the server URL. Once connected, call `client.get_tools()` and pass them directly to your agent constructor to start tracking runs.

Yes. LangChain agents use the ReAct framework to decide when to call `list_experiments` and when to pull specific metrics based on what they find in your workspaces.

LangSmith captures the exact inputs and outputs of every tool call, like `get_experiment_metrics`. You can trace latency, see token usage, and inspect the raw payloads returned from your Comet workspace.

Absolutely. You can mix these machine learning tools with database or vector store tools, allowing your agent to pull training metrics and write a summary report to your database in one run.

Your credentials never touch the LLM or LangChain. The MCP Server runs in a secure sandbox on Vinkius, meaning your workspace metadata and training telemetry are kept isolated and safe.

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