How to Use the HubSpot CRM (Full) MCP in LangChain

Q: How do I connect HubSpot CRM (Full) to LangChain?

Run pip install langchain-mcp-adapters langgraph. Wrap the endpoint URL in MultiServerMCPClient and call gettools() to pass the functions to your agent.

Q: Can LangChain agents read existing deal amounts?

Yes. The hssearchdeals tool returns the monetary amount, pipeline stage, and expected close date. Your agent can read this data before deciding to update the record.

Q: Does this support LangGraph state persistence?

It does. You call client.session() to maintain persistent context across multiple turns. The agent remembers the pipeline ID it retrieved five steps ago.

Build ReAct agents in LangChain that read, write, and trace HubSpot CRM records directly from your Python pipelines.

See Vinkius in Action

Works with every AI agent you already use

…and any MCP-compatible client

MCP Servers - Free for Subscribers

Connect HubSpot CRM (Full) MCP to LangChain

Create your Vinkius account to connect HubSpot CRM (Full) 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 HubSpot CRM (Full) with LangChain

Ask AI about this MCP

ChatGPT

Claude

Perplexity

Sequential deal generation

The `hs_list_pipelines` tool feeds required stage IDs directly into `hs_create_deal` within your ReAct loops. Your agent grabs the exact display order for your sales pipeline, picks the right stage, and generates the opportunity with a required dealname. Stop hardcoding pipeline structures. LangChain passes the output from the list operation straight into the creation tool. That means your agent dynamically adapts if Sales Ops adds a new Proposal stage tomorrow.

Agentic contact enrichment

Calling `hs_search_companies` lets your agent find an organization's domain before it triggers `hs_create_contact`. This exact sequence prevents duplicate entries. The agent verifies the company exists, pulls the assigned owner, and associates the new email address with the correct account. LangGraph excels at this kind of conditional logic. If the search returns empty, the chain halts. When it finds a match, the agent sets the lifecycle stage to salesqualifiedlead and logs the new contact ID.

Full LangChain MCP Server tracing

Executing `hs_search_tickets` and `hs_create_note` generates complete execution graphs in LangSmith. You see exactly what subject the agent queried and which ticket IDs it updated. Every parameter sent to the MCP Server gets logged. Debugging complex workflows is dead simple. When an agent assigns a ticket to the wrong person, you check the trace for `hs_list_owners`. The input and output payloads tell you instantly if the agent hallucinated a name or if the API returned stale data.

Setup guide

Set up HubSpot CRM (Full) 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 HubSpot CRM (Full) 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({
    "hubspot-crm-full-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 HubSpot CRM (Full) 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 HubSpot. 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 HubSpot CRM (Full) 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 HubSpot CRM (Full) MCP in LangChain

Run `pip install langchain-mcp-adapters langgraph`. Wrap the endpoint URL in `MultiServerMCPClient` and call `get_tools()` to pass the functions to your agent.

Yes. The `hs_search_deals` tool returns the monetary amount, pipeline stage, and expected close date. Your agent can read this data before deciding to update the record.

HubSpot auto-deduplicates based on the unique email address provided to `hs_create_contact`. The agent receives the existing contact ID instead of creating a messy duplicate.

It does. You call `client.session()` to maintain persistent context across multiple turns. The agent remembers the pipeline ID it retrieved five steps ago.

We run a V8 Isolate Sandbox with zero-trust architecture. When your agent queries customer phone numbers and emails via `hs_search_contacts`, that memory space is ephemeral and destroyed immediately after the execution completes.

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