Firecrawl MCP Server

Official

Firecrawl's official MCP server for web search, scraping, and structured extraction for AI agents

How to install ↓ Compare GitHub source repository ↗

Category: Other #2 of 230
Stars: ★ 7.6k Very popular
Transport: stdio (local process)
Runtime: Node.js · Hosted (no local runtime)
Credentials: Optional API key
License: MIT
Last commit: 4 days ago
Tools: 11

75FMRS · B

Officially maintained by Firecrawl; tools cover the full range from single-page scraping to site-wide crawling, structured extraction, and an interactive agent, with usage guidance and common pitfalls documented per tool in the README. This review is a static check of source and docs, not an actual call test.

Strongest · Setup experience 17/20 Weakest · Security and permissions 12/20

Reliability

13/20

Security and permissions

12/20

Maintenance

16/20

Documentation

17/20

Setup experience

17/20

Test out of dateListed 2026-07-24

Read the FMRS scoring method →

Fit and risk

What it can accessUses the networkControls a browser

Best for

Scenarios needing clean, structured web context fed to an AI assistant
Users already on the Firecrawl platform who want to call its scraping capability directly via MCP

Not for

Simple static-page scraping where you don't want to depend on a third-party API and incur call costs
Sites that explicitly disallow automated access (robots.txt)

Required permissions

Requires a Firecrawl API key to call; cost and quota are governed by the Firecrawl account
firecrawl_agent/firecrawl_interact perform automated browser interaction, which may trigger login or form-submission flows on the target site

Risks and side effects

Bulk crawl/map tools can generate significant request volume against a target site — respect the site's rate limits and terms of service
firecrawl_agent's interactive action chain is longer — define task boundaries clearly before running it to avoid accidentally triggering actions on a sensitive site

Setup

Before you start

Runtime：Node.js · Hosted (no local runtime)

FIRECRAWL_API_KEY requiredsecret API key for the Firecrawl cloud service; get one at https://www.firecrawl.dev/app/api-keys. Optional when using a self-hosted instance.

FIRECRAWL_OAUTH_TOKEN optionalsecret Static OAuth access token (fco_ prefix) issued by the hosted service, used for stdio-mode bearer authentication.

Other optional settings (8)

FIRECRAWL_API_URL optional Custom endpoint for a self-hosted Firecrawl instance, e.g. https://firecrawl.your-domain.com; defaults to the cloud API.

HTTP_STREAMABLE_SERVER optional Set to true to run the server in local Streamable HTTP mode at http://localhost:3000/mcp.

CLOUD_SERVICE optional Set to true to enable the hosted HTTP stream transport mode.

SSE_LOCAL optional Set to true to enable local SSE transport.

FIRECRAWL_NO_SEARCH_FEEDBACK optional Set to 1 to prevent the firecrawl_search_feedback tool from being registered.

FIRECRAWL_DISABLE_SEARCH_FEEDBACK optional Alias of FIRECRAWL_NO_SEARCH_FEEDBACK, also disables the search feedback tool.

FIRECRAWL_NO_ENDPOINT_FEEDBACK optional Set to 1 to prevent the firecrawl_feedback tool from being registered.

FIRECRAWL_DISABLE_ENDPOINT_FEEDBACK optional Alias of FIRECRAWL_NO_ENDPOINT_FEEDBACK, also disables the endpoint feedback tool.

Start via npx -y firecrawl-mcp, after signing up at https://www.firecrawl.dev and setting a FIRECRAWL_API_KEY env var. A self-hosted Firecrawl instance can also be pointed to via the corresponding env var.

claude_desktop_config.json

{
  "mcpServers": {
    "firecrawl": {
      "command": "npx",
      "args": ["-y", "firecrawl-mcp"],
      "env": {
        "FIRECRAWL_API_KEY": "<YOUR_FIRECRAWL_API_KEY>"
      }
    }
  }
}

Shown for Claude Desktop. Other clients may use a different file or key (VS Code uses "servers") — the configurator below converts it.

.vscode/mcp.json

{
  "servers": {
    "firecrawl": {
      "command": "npx",
      "args": [
        "-y",
        "firecrawl-mcp"
      ],
      "env": {
        "FIRECRAWL_API_KEY": "<YOUR_FIRECRAWL_API_KEY>"
      }
    }
  }
}

Goes in your project's .vscode/mcp.json (VS Code uses a "servers" key).

Terminal

claude mcp add firecrawl -e 'FIRECRAWL_API_KEY=<YOUR_FIRECRAWL_API_KEY>' -- npx -y firecrawl-mcp

Run it in a terminal; replace any <…> placeholders with your own values first.

Generate and validate client config

Check that it works

After refreshing the MCP server list in your client, tools like firecrawl_scrape and firecrawl_search should appear; then issue a simple scrape request — a returned page result confirms the connection works.

Troubleshooting

Calls return an auth error: check that FIRECRAWL_API_KEY is correct and the account has remaining quota
A crawl job appears stuck: use firecrawl_check_crawl_status to check progress — very large sites can take a while
Extraction returns empty: confirm the schema passed to firecrawl_extract matches the page's actual structure

Things to try

Once connected, you can ask your AI assistant things like:

Extract the product name and price from https://example.com/product using JSON format
List all URLs on example.com
Search for the latest AI research papers published in 2023
Open https://example.com, click the pricing link and summarize the visible plans

Tools 11

firecrawl_scrape read-only

Scrape a single URL's content with advanced format options

firecrawl_map read-only

Discover all crawlable URLs under a website

firecrawl_search read-only

Search the web and return relevant page content

firecrawl_crawl read-only

Deep-crawl multiple pages across an entire website

firecrawl_check_crawl_status read-only

Check the progress and status of a crawl job

firecrawl_parse read-only

Parse already-fetched content into a structured form

firecrawl_extract read-only

Extract structured data from a page against a given schema

firecrawl_agent writes

Run a browser agent to complete complex, interactive web tasks

Show 3 more tools

firecrawl_agent_status read-only

Check the execution status of an agent task

firecrawl_interact writes

Interact with a webpage (click, type, etc.)

firecrawl_interact_stop writes

Stop an in-progress web interaction session

Use cases

Scrape a single page's content for summarization or Q&A

Map and deep-crawl an entire site to gather content in bulk

Extract structured data (e.g. product info, pricing) from a page against a schema

Use the browser agent to complete complex scraping tasks requiring clicks or logins

Supported clients

Claude Desktop

VS Code

Cursor

Windsurf

Zed

Amp

Listed from the project's documentation, not tested by this site.

Overview

The official Firecrawl-maintained MCP server, exposing the Firecrawl platform's page scraping, site-map discovery, deep crawling, structured data extraction, and an interactive browser agent to MCP-compatible AI assistants — for getting clean, ready-to-use web context.

Similar servers

Hevy MCP 77 · B

Manage your Hevy workout data from AI assistants via MCP.

★ 517 · Tools 26 Compare with this →

BioMCP 75 · B

One binary. One grammar. Evidence from the biomedical sources you already trust.

★ 659 · Tools 11 Compare with this →

Compartment 75 · B

Encrypted, fully offline long-term memory for AI agents

★ 583 · Tools 14 Compare with this →

Exa MCP Server 73 · B

Exa's official MCP server for web search and crawling built for AI applications

★ 5.1k · Tools 4 Compare with this →

Compare with other MCP servers

Source revision 0e3bb9c0f8bc Data synced 2026-10-19 Read the FMRS scoring method