How to Use the Firecrawl MCP in LangChain

Q: Can a LangChain chain monitor long-running Firecrawl crawl jobs?

Yes, by using a loop in your chain that queries getcrawlstatus. Your LangChain agent can check the state periodically and only proceed to the next chain step once the job is marked complete.

Q: How do I handle rate limits when using Firecrawl with LangChain?

You should monitor your credit consumption by calling getapiusage within your chain logic. If your LangChain run detects that you are nearing your limit, you can trigger cancelactivecrawl to stop active background jobs.

Feed clean markdown from any URL directly into your LangChain chains with the Firecrawl 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 Firecrawl MCP to LangChain

Create your Vinkius account to connect Firecrawl 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 Firecrawl with LangChain

Ask AI about this MCP

ChatGPT

Claude

Perplexity

Build multi-step LangChain scraping pipelines

LangChain agents can coordinate complex web crawls by chaining Firecrawl tools. Your agent can first call `map_website_structure` to find target URLs, and then dynamically decide whether to trigger `start_crawl` for bulk extraction or run `scrape_url` for a quick single-page fetch. Because LangChain handles state transitions between chain links, the output of one Firecrawl tool flows directly into the next. If a bulk job takes too long, your agent can monitor progress with `get_crawl_status` and execute `cancel_active_crawl` to prevent unnecessary token consumption.

Track scrape costs inside LangSmith

Every single Firecrawl tool call made by your LangChain agent gets fully logged in LangSmith. You see exactly how many credits were spent on JS-rendering by checking the response payload of `get_api_usage` right alongside your agent's reasoning traces. This visibility prevents runaway costs when running deep recursive crawls. Instead of guessing why your API bill is spiking, you can inspect the exact inputs passed to `start_crawl` and fine-tune your agent's prompt to target fewer pages.

Convert raw web pages into structured chain inputs

Stop writing custom regex or BeautifulSoup parsers for your LangChain document loaders. When your agent invokes `scrape_url`, the Firecrawl MCP Server returns clean, structured markdown that fits perfectly into your LangChain prompt templates. This means your chains receive pure content without the surrounding navigation menus, footer links, or cookie banners. Your model gets straight to the core data, reducing prompt token waste and keeping context windows clean.

Setup guide

Set up Firecrawl 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 Firecrawl 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({
    "firecrawl-extended-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 Firecrawl 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 Firecrawl. 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 Firecrawl 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 Firecrawl MCP in LangChain

You load the server using the LangChain MCP adapter and get the tool list. Your agent then calls `map_website_structure` or `scrape_url` directly inside its execution loop, converting the markdown output into system messages.

Yes, by using a loop in your chain that queries `get_crawl_status`. Your LangChain agent can check the state periodically and only proceed to the next chain step once the job is marked complete.

You should monitor your credit consumption by calling `get_api_usage` within your chain logic. If your LangChain run detects that you are nearing your limit, you can trigger `cancel_active_crawl` to stop active background jobs.

Yes, the server handles heavy JS hydration automatically when you call `scrape_url`. Your agent receives the fully rendered markdown without needing to configure headless browsers or proxy networks.

All scraped HTML and markdown content fetched from target URLs goes through ephemeral Vinkius sandboxes. No scraped data is stored on our servers after the tool execution 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