How to Use the Comet ML MCP in AutoGen

Q: How do I register the Comet ML MCP Server tools with AutoGen?

Use the autogen-extmcp package to connect to the server. Pass the server parameters to mcpservertools and assign the generated tools list directly to your AssistantAgent.

Q: Can multiple AutoGen agents share access to Comet ML tools?

Yes. You can pass the same tools, like getexperimentparams, to multiple agents so they can collaborate, debate, and double-check each other's analysis of your training runs.

Let AutoGen agents debate and analyze Comet ML metrics to make deployment decisions using 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 Comet ML MCP to AutoGen

Create your Vinkius account to connect Comet ML to AutoGen 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 AutoGen

Ask AI about this MCP

ChatGPT

Claude

Perplexity

Multi-agent workspace auditing in AutoGen

This tool lets your agents call `list_workspaces` to identify active environments and audit them for compliance. One agent calls the workspace list, while another uses `list_projects` to check for naming standards and project structures. They discuss their findings in a shared conversation thread, flagging projects that do not match your team's guidelines. This automated peer review keeps your Comet workspaces clean without manual checking.

Debate model deployment readiness using Comet ML metrics

This tool lets your agents call `get_experiment_metrics` to check numeric bounds and debate whether a model is ready for production. A performance agent checks the metrics, while a safety agent analyzes the run parameters. This MCP Server integration lets your agents negotiate based on the live data, deciding whether the model meets your performance and safety thresholds. This consensus-driven approach ensures no model is deployed without meeting all criteria.

Automate experiment troubleshooting with AutoGen

This tool lets your agents call `list_experiments` to retrieve the run list and coordinate troubleshooting of failed runs. One agent retrieves the run list, while a debugger agent pulls the exact payload using `get_experiment` to inspect the failure. They collaborate to isolate the bad hyperparameters by comparing the failed run against successful ones. This cooperative debugging cuts down the time you spend digging through training logs.

Setup guide

Set up Comet ML MCP in AutoGen

Prerequisites

Python 3.10+ installed
autogen-ext[mcp] package
Active Vinkius subscription with a valid endpoint token

1

Install AutoGen with MCP
Run pip install "autogen-ext[mcp]" autogen-agentchat. The MCP extension includes mcp_server_tools for stateless tool access.
2

Fetch tools from the MCP
Call mcp_server_tools(SseServerParams(url=...)) with your Vinkius endpoint. Replace [YOUR_TOKEN_HERE] with your token from cloud.vinkius.com.
3

Run your agent
Pass the tools to AssistantAgent and call agent.run(). The agent invokes Comet ML tools and returns structured results.

agent.py

from autogen_ext.tools.mcp import SseServerParams, mcp_server_tools
from autogen_agentchat.agents import AssistantAgent
from autogen_ext.models.openai import OpenAIChatCompletionClient

server_params = SseServerParams(
    url="https://edge.vinkius.com/[YOUR_TOKEN_HERE]/mcp"
)

tools = await mcp_server_tools(server_params)

agent = AssistantAgent(
    name="Comet ML_assistant",
    model_client=OpenAIChatCompletionClient(model="gpt-4o"),
    tools=tools,
)

result = await agent.run("List recent Comet ML data")
print(result.messages[-1].content)

Get your connection token →

Prerequisites

Python 3.10+ installed
autogen-ext[mcp] + autogen-agentchat
Active Vinkius subscription with a valid endpoint token

1

Install dependencies
Same packages as above. McpWorkbench is ideal when your agent needs stateful sessions across multiple tool calls.
2

Use McpWorkbench as context manager
Wrap your agent in async with McpWorkbench(...) to maintain shared state and resources. The workbench manages the full MCP session lifecycle.
3

Run with workbench
Pass workbench=workbench to your agent. State is preserved across multiple tool calls within the same session.

agent.py

from autogen_ext.tools.mcp import McpWorkbench, SseServerParams
from autogen_agentchat.agents import AssistantAgent
from autogen_ext.models.openai import OpenAIChatCompletionClient

server_params = SseServerParams(
    url="https://edge.vinkius.com/[YOUR_TOKEN_HERE]/mcp"
)

async with McpWorkbench(server_params) as workbench:
    agent = AssistantAgent(
        name="Comet ML_assistant",
        model_client=OpenAIChatCompletionClient(model="gpt-4o"),
        workbench=workbench,
    )

    result = await agent.run("List recent Comet ML data")
    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 AutoGen

Use the `autogen-ext[mcp]` package to connect to the server. Pass the server parameters to `mcp_server_tools` and assign the generated tools list directly to your `AssistantAgent`.

Yes. You can pass the same tools, like `get_experiment_params`, to multiple agents so they can collaborate, debate, and double-check each other's analysis of your training runs.

They negotiate in a conversation loop. The performance agent reads `get_experiment_metrics` and argues for deployment, while the quality agent checks the parameters and can veto if limits are exceeded.

Yes. The adapter handles both stdio and Streamable HTTP transports, allowing your agents to communicate with the server regardless of your hosting setup.

The MCP Server runs inside a zero-trust, ephemeral sandbox. Your model parameters and training logs are fetched on demand and never cached or exposed to third parties.

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