How to Use the ArcGIS MCP in LangChain

Q: How do I handle batch geocoding errors in an ArcGIS LangChain pipeline?

Use an exception-handling chain wrapper. When batchgeocodeaddresses throws an error due to bad formatting, your LangChain agent can catch the exception, parse the raw input, and attempt to clean it using a fallback tool before retrying.

Q: How does the ArcGIS MCP Server optimize travel time queries in LangChain?

It exposes calculateodmatrix directly to your agent. Instead of writing custom API integration code, your LangChain agent invokes the tool to get a full distance matrix in a single step, saving API overhead and reducing token usage.

Build spatial reasoning chains in LangChain using live ArcGIS location data.

See Vinkius in Action

Works with every AI agent you already use

…and any MCP-compatible client

MCP Servers - Free for Subscribers

Connect ArcGIS MCP to LangChain

Create your Vinkius account to connect ArcGIS 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 ArcGIS with LangChain

Ask AI about this MCP

ChatGPT

Claude

Perplexity

Chain Multi-Stop Routes with LangChain

Connect your LangChain agents to real-world logistics. By exposing the `solve_vehicle_routing` tool on this MCP Server, your chains can calculate complex fleet schedules and pass those direct coordinates straight into `solve_nav_route` for turn-by-turn directions. Since LangChain manages state across multi-step runs, your agent can inspect the output of a route calculation, check for anomalies, and re-run the chain with adjusted parameters without manual code. It makes spatial logistics a native step in your agentic workflows.

Debug ArcGIS MCP Server Tasks inside LangSmith

Stop guessing why a spatial query failed. When your agent calls `calculate_od_matrix` or `calculate_service_area`, LangSmith captures the exact payload, latency, and token cost of that specific MCP Server tool execution. This transparency lets you pinpoint exactly where a geocoding chain went sideways. You see the raw coordinates passed to the server and the exact matrix returned, making performance tuning incredibly fast.

Intelligent Geocoding with ReAct Loops

Raw text addresses are notoriously messy. Your LangChain agent can use `suggest_geocoding` first to clean up user input, then feed the validated suggestion directly into `find_address_candidates` to get clean, structured coordinates. If the initial lookup returns multiple possibilities, the ReAct loop evaluates the candidates against your business logic. It chooses the best match or asks the user for clarification, handling the entire validation pipeline autonomously.

Setup guide

Set up ArcGIS 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 ArcGIS 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({
    "arcgis-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 ArcGIS 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 ArcGIS. 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 ArcGIS 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 ArcGIS MCP in LangChain

Use an exception-handling chain wrapper. When `batch_geocode_addresses` throws an error due to bad formatting, your LangChain agent can catch the exception, parse the raw input, and attempt to clean it using a fallback tool before retrying.

Yes, by using LangGraph or a conversational agent. The agent stores previous stops from `solve_nav_route` in its thread history, allowing it to calculate a return path or append a new destination to the existing itinerary in subsequent steps.

It exposes `calculate_od_matrix` directly to your agent. Instead of writing custom API integration code, your LangChain agent invokes the tool to get a full distance matrix in a single step, saving API overhead and reducing token usage.

Install langchain-mcp-adapters and run the MultiServerMCPClient pointing to your Vinkius endpoint. From there, extract the tools and pass them directly to your agent initialization block.

Yes, it is. Vinkius runs the server inside a secure V8 Isolate Sandbox. Your address strings and coordinates are processed in an ephemeral environment, meaning no location logs are retained once the API call 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