How to Use the Helicone (LLM Observability) MCP in LangChain

Q: Can my LangChain agent automatically check API costs with Helicone (LLM Observability)?

Yes, your agent can call querycosts directly mid-chain. This allows you to build guardrails that stop execution if a run gets too expensive.

Q: How does Helicone (LLM Observability) help LangChain handle slow model responses?

The agent uses querylatency to monitor response times. If a model slows down, the chain can automatically route subsequent requests to a faster backup model.

Q: Can I track user feedback directly in my chains?

Yes, you can. Use logfeedback to push user ratings back to Helicone, then query that data later using queryfeedback to see which chains perform best.

Track your LangChain chain costs and latencies in real time with the Helicone 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 Helicone (LLM Observability) MCP to LangChain

Create your Vinkius account to connect Helicone (LLM Observability) 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 Helicone (LLM Observability) with LangChain

Ask AI about this MCP

ChatGPT

Claude

Perplexity

Trace LangChain token spend using the MCP Server

Running `query_costs` helps you stop flying blind on model expenses when your multi-step LangChain agents execute complex reasoning paths. It's too easy to run up a massive bill without realizing it until the invoice hits. You can also map custom metadata with `list_properties` to see which specific prompts or users are draining your budget. It gives you the raw numbers directly inside your execution graph.

Isolate slow steps in your chains

Triggering `query_latency` allows you to isolate and fix slow steps in your chains before they kill the user experience. If your LangChain agent gets stuck in a loop, you need to know exactly which model or tool caused the backup. Your chains can dynamically swap to faster models if latency spikes. By combining this with `query_requests`, you get a clear picture of slow execution paths without digging through raw log files.

Debug prompt versions in production

Executing `get_prompt_versions` lets you debug prompt changes directly within your chain when a new deployment degrades performance. Your LangChain pipeline needs to pull the exact history to see what changed. You can pair this with `query_prompts` to inspect the actual inputs that hit your models. It keeps your prompt engineering grounded in real production data instead of guesswork.

Setup guide

Set up Helicone (LLM Observability) 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 Helicone (LLM Observability) 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({
    "helicone-llm-observability-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 Helicone (LLM Observability) 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 Helicone. 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 Helicone (LLM Observability) 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 Helicone (LLM Observability) MCP in LangChain

You connect the MCP Server to your LangChain setup using the HTTP adapter. The agent can then query metrics like `query_sessions` to trace step-by-step execution across your entire pipeline.

Yes, your agent can call `query_costs` directly mid-chain. This allows you to build guardrails that stop execution if a run gets too expensive.

The agent uses `query_latency` to monitor response times. If a model slows down, the chain can automatically route subsequent requests to a faster backup model.

Yes, you can. Use `log_feedback` to push user ratings back to Helicone, then query that data later using `query_feedback` to see which chains perform best.

Your raw prompt text and token costs are handled securely. Vinkius runs the server in an isolated sandbox, meaning your API keys and raw logs are never stored 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