How to Use the Exa AI MCP in LangChain

Q: How do I pass Exa AI search results to another LangChain tool?

You use the output of searchweb directly as the input for your next chain node. Because the server returns structured JSON, your agent can parse the links and immediately pass them to your database writers or summarize chains.

Q: Can I trace Exa AI tool calls inside LangSmith with LangChain?

Yes, every time your agent calls getcontents or findsimilar, LangSmith logs the exact input parameters and response times. This makes it easy to debug why a specific search step failed or took too long.

Q: How do I prevent my agent from scraping too much data?

Use getsearchlinks instead of fetching full pages. This tool returns only the top URLs, allowing your agent to filter the links first before deciding which ones are worth pulling down with getcontents.

Feed clean, semantic search results directly into your LangChain runs with the Exa AI 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 Exa AI MCP to LangChain

Create your Vinkius account to connect Exa AI 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 AI with LangChain

Ask AI about this MCP

ChatGPT

Claude

Perplexity

Build multi-step LangChain search loops with this MCP Server

Running `search_with_contents` allows your LangChain agent to execute deep web research without raw HTML noise. The tool grabs semantic results and instantly feeds that clean text into the next step of your chain. You don't have to write custom scrapers or regex parsers; you just get the data you actually need. Every step of this process is fully visible in your LangSmith dashboard. When the agent uses `find_similar` to expand its search space, you can trace the latency and token usage of each individual tool call. This lets you optimize your chains so they run faster and cost less.

Chain semantic search directly into your document loaders

Executing `search_web` lets your LangChain agent fetch fresh, live web links on the fly. This MCP Server lets you plug these results straight into your existing agentic workflows to ground decisions in real-time facts instead of outdated static data. If you need the full picture, have the chain trigger `get_contents` using the specific document IDs returned from your search. This fetches clean text, stripped of ads and navigation links, so you don't waste tokens on junk.

Track live crawl usage and API costs in your pipelines

Exposing `get_api_usage` inside your LangChain runs gives your agent direct access to its own crawl statistics before making more calls. This prevents complex ReAct agents from looping endlessly on search queries and running up massive bills. Your pipeline gains self-monitoring capabilities this way. The agent can switch to `get_search_links` for quick, cheap URL lists if it detects that it is getting close to its API limits during a long-running batch job.

Setup guide

Set up Exa AI 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 AI 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-ai-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 AI 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 AI. 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 AI 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 AI MCP in LangChain

You use the output of `search_web` directly as the input for your next chain node. Because the server returns structured JSON, your agent can parse the links and immediately pass them to your database writers or summarize chains.

Yes, every time your agent calls `get_contents` or `find_similar`, LangSmith logs the exact input parameters and response times. This makes it easy to debug why a specific search step failed or took too long.

Yes, you can load this server alongside your database tools using the MultiServerMCPClient. This lets your LangChain agent query the web and write the findings to a local database in a single execution loop.

Use `get_search_links` instead of fetching full pages. This tool returns only the top URLs, allowing your agent to filter the links first before deciding which ones are worth pulling down with `get_contents`.

Your API usage data and crawl statistics never touch third-party logging platforms. This MCP integration runs inside an isolated sandbox, meaning your Exa API keys and search metrics stay strictly local to your environment.

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