How to Use the Granola MCP in LangChain

Q: How do I connect the Granola MCP Server to my LangChain project?

Install langchain-mcp-adapters and langgraph via pip to handle the MCP transport. Initialize the MultiServerMCPClient with your Vinkius endpoint URL, pull the tools with gettools(), and pass them to your ReAct agent.

Q: Can a LangChain agent search across all my Granola documents?

Yes, your agent doesn't need to guess—it calls searchdocuments to run full-text queries across your entire meeting history. It can then feed the results directly into other LangChain integrations.

Q: How does LangChain handle large meeting transcripts from Granola?

The agent uses gettranscript to retrieve the text with speaker detection, then passes it through a LangChain splitter. This prevents token overflow while keeping the speaker context intact.

Q: Can I trace Granola tool execution in LangSmith?

Every tool call, including getcontent and getparticipants, is tracked automatically in LangSmith. You see exactly how long the tool took and the exact JSON payload returned.

Run LangChain ReAct agents that pull meeting notes, transcripts, and action items directly into your composable multi-step chains.

See Vinkius in Action

Works with every AI agent you already use

…and any MCP-compatible client

MCP Servers - Free for Subscribers

Connect Granola MCP to LangChain

Create your Vinkius account to connect Granola 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 Granola with LangChain

Ask AI about this MCP

ChatGPT

Claude

Perplexity

Chained Meeting Analysis in LangChain

This MCP Server exposes tools like `get_transcript` and `get_summary` so your LangChain agents can analyze discussions in a structured pipeline. The output of your transcript retrieval feeds directly into the next chain step, letting the agent process raw meeting text without manual context passing. You can track the latency and token cost of these multi-step operations using LangSmith. If the agent needs to verify a specific detail, it calls `get_metadata` to confirm the meeting date before running a deeper analysis.

Targeted Action Item Extraction

The `get_action_items` tool extracts clear, owner-assigned tasks directly from your Granola meeting documents. Your agent calls this tool to isolate what needs to get done, then immediately forwards those items to databases or task managers in the same chain. For complex follow-ups, the agent can chain `get_documents_batch` to aggregate tasks across multiple sessions. This keeps your task lists grounded in the actual meeting text rather than guesses.

Chronological Workspace Auditing

The `list_by_date` tool lets your agent scan Granola documents within specific timeframes to build historical reports. Your agent uses this chronological data to trace how decisions evolved over consecutive project syncs. By combining this with `list_recent` and `list_folders`, the agent navigates your entire workspace structure programmatically. It filters down to the exact meeting notes required before executing deeper summary prompts.

Setup guide

Set up Granola 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 Granola 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({
    "granola-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 Granola 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 Granola. 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 Granola 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 Granola MCP in LangChain

Install `langchain-mcp-adapters` and `langgraph` via pip to handle the MCP transport. Initialize the `MultiServerMCPClient` with your Vinkius endpoint URL, pull the tools with `get_tools()`, and pass them to your ReAct agent.

Yes, your agent doesn't need to guess—it calls `search_documents` to run full-text queries across your entire meeting history. It can then feed the results directly into other LangChain integrations.

The agent uses `get_transcript` to retrieve the text with speaker detection, then passes it through a LangChain splitter. This prevents token overflow while keeping the speaker context intact.

Every tool call, including `get_content` and `get_participants`, is tracked automatically in LangSmith. You see exactly how long the tool took and the exact JSON payload returned.

Your meeting transcripts and action items remain locked in Vinkius's zero-trust sandboxed environment. This MCP integration keeps your data locked down, fetching document data only when your LangChain agent explicitly invokes tools like `get_content`.

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