How to Use the Exa MCP in LangChain

Q: Can I use Exa to find similar technical docs in a LangChain chain?

Yes, you can use the findsimilar MCP tool within your chain to locate related pages. Your LangChain agent can take a high-performing URL from a previous step and feed it into this tool to expand its research base.

Q: How do I monitor Exa tool calls made by my LangChain agent?

Every tool call, including searchneural and getcontents, automatically registers with LangSmith if you have tracing enabled. You can inspect the exact query strings, response payloads, and execution times for every single search.

Q: What is the difference between keyword and neural search tools?

The searchkeyword tool looks for exact word matches, which is great for error codes or specific names. The searchneural tool uses deep learning to find pages based on the meaning of your query, even if the exact words aren't there.

Feed clean, context-rich web search data directly into your LangChain reasoning loops without writing custom scrapers.

See Vinkius in Action

Works with every AI agent you already use

…and any MCP-compatible client

MCP Servers - Free for Subscribers

Connect Exa MCP to LangChain

Create your Vinkius account to connect Exa 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 Exa with LangChain

Ask AI about this MCP

ChatGPT

Claude

Perplexity

Multi-step search routing with LangChain agents

The `search_neural` tool lets your LangChain agents run conceptual web searches that understand context instead of just matching keywords. By exposing this to your ReAct loop, your agent can start with a broad query to grasp a concept, then immediately pivot. You don't need to hardcode API endpoints or write custom JSON parsers. LangSmith tracks every tool call so you can see exactly which tool was called and why.

Target specific domains inside your chains

The `search_domain` tool restricts your LangChain agent's searches to specific, trusted websites. This keeps your agent from wandering off into random forums when it should be reading official documentation. If the agent needs to compare pages, it swaps to `find_similar` to discover related resources within those same domains. This targeted approach keeps your token usage low and prevents your chain from getting clogged with irrelevant noise.

Real-time data extraction for LangChain pipelines

The `search_with_contents` tool retrieves both the search results and the clean page text in a single step for your LangChain chains. Your chain gets the full context immediately, cutting down on extra LLM roundtrips. For fast-moving topics, the agent can run `search_recent` to grab the latest news. It then pipes that fresh content directly into your synthesis chain, ensuring your output isn't using stale information from last year.

Setup guide

Set up Exa 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 Exa 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({
    "exa-alternative-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 Exa 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 Exa. 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 Exa 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 Exa MCP in LangChain

The Exa MCP Server executes searches over highly optimized endpoints, returning clean text in sub-second times. This prevents your LangChain agent from stalling during complex, multi-step search and retrieval loops.

Yes, you can use the `find_similar` MCP tool within your chain to locate related pages. Your LangChain agent can take a high-performing URL from a previous step and feed it into this tool to expand its research base.

Every tool call, including `search_neural` and `get_contents`, automatically registers with LangSmith if you have tracing enabled. You can inspect the exact query strings, response payloads, and execution times for every single search.

The `search_keyword` tool looks for exact word matches, which is great for error codes or specific names. The `search_neural` tool uses deep learning to find pages based on the meaning of your query, even if the exact words aren't there.

Your web search queries and retrieved page contents are processed in an isolated Vinkius sandbox. We do not store or log your queries, and your API keys are protected using zero-trust environment variables.

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