How to Use the EOSDA MCP in LangChain

Q: Can my LangChain agent use EOSDA to compare crop health across multiple fields?

Yes, that's a perfect use case. The agent can first call getfields to get all your field IDs. Then, it can loop through those IDs, calling getndvitimeseries for each one and comparing the results to find underperforming areas.

Build autonomous ag-data pipelines with LangChain and the EOSDA 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 EOSDA MCP to LangChain

Create your Vinkius account to connect EOSDA 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 EOSDA with LangChain

Ask AI about this MCP

ChatGPT

Claude

Perplexity

Chain Satellite Data to Decisions

Connect EOSDA tools into a logical sequence that your LangChain agent executes automatically. Start by getting a list of fields with `get_fields`, then pass a field ID to `get_ndvi_timeseries` to check its health trend over the last 90 days. If the agent spots a problem, it can automatically generate a high-contrast visualization using `render_index_map` for a human to review. This isn't just calling a single tool. It's about building a repeatable process where the output of one step feeds the next. Your agent gets smarter, handling multi-step requests like, "Check all my corn fields for water stress and show me the worst one." The agent decides the right sequence of tools to call—`get_fields`, `get_ndmi_timeseries`, and `render_index_map`—to get you the answer.

Full Observability with LangChain

Every call your agent makes to the EOSDA MCP Server is visible. When you build a chain to analyze field conditions, you get a complete trace. You see the exact GeoJSON sent to `create_field`, the raw soil moisture stats returned by `get_soil_moisture`, and the latency of each step. This makes debugging a hundred times easier. If a chain produces a weird result, you don't have to guess why. The trace shows you the agent's reasoning, the data it used from `get_weather_data`, and where it went wrong. It's the only way to build production-grade agents you can actually trust.

Create Complex Agricultural Workflows

Go beyond simple Q&A. Use LangChain to construct sophisticated workflows that mirror real-world farm management. For example, build an agent that runs weekly: it pulls the `get_weather_forecast`, checks `get_soil_moisture` levels, and decides if irrigation is needed for the upcoming week. Combine multiple data points for a single, informed decision. Your agent can cross-reference `get_evi_timeseries` with historical `get_weather_data` to predict yield potential or flag anomalies that a single index might miss. This MCP server gives your agent the raw data; LangChain provides the structure to reason with it.

Setup guide

Set up EOSDA 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 EOSDA 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({
    "eosda-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 EOSDA 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 EOSDA. 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 EOSDA 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 EOSDA MCP in LangChain

You instantiate the MCP client from the adapter library and call `get_tools()`. Pass that tool list directly to your agent constructor. LangChain handles the rest, making tools like `get_fields` and `get_zoning_map` available to your agent's reasoning loop.

Yes, that's a perfect use case. The agent can first call `get_fields` to get all your field IDs. Then, it can loop through those IDs, calling `get_ndvi_timeseries` for each one and comparing the results to find underperforming areas.

A good approach is to have the agent gather data with multiple tool calls—like `get_soil_moisture`, `get_weather_forecast`, and `get_vegetation_index`. Then, it can feed that structured data into another LLM call to synthesize a summary report.

It grounds your agent in facts. Instead of guessing about weather or crop health, your agent can get hard numbers by calling `get_weather_data` or `get_ndvi_timeseries`. The answers it gives are based on real-time satellite data, not just its training.

Yes. The MCP Server itself is ephemeral and doesn't store your data. When your LangChain agent sends a GeoJSON polygon to `create_field`, it passes through our Vinkius sandbox directly to the EOSDA API. The only place that data might be logged is in your own LangSmith traces, which you control.

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