Name: Crawleo MCP Server
Author: Crawleo

LABSCO SUMMARY

It answers the format question that usually costs an extra step.

Both tools return the body in the same three shapes — markdown, raw HTML and enhanced HTML — so a page arrives ready for a model to read rather than needing conversion afterwards. The search tool is the one carrying locale: cc, geolocation, setLang and device change what comes back, and max_pages decides how far it goes. With 2 tools there is nothing else in the surface: no crawl of a whole site, no queue, no extraction schema.

What it is

An MCP server in front of Crawleo's search and crawl service.

What you get

Web search returning results with page text and structured data, shaped by country, geolocation, language and device, and extended across more than one page with max_pages.
A single URL crawled and returned as markdown, raw HTML or enhanced HTML — the same body formats the search side offers.

Requirements

CRAWLEO_API_KEY, since both tools run against Crawleo's own service.

Setup effort

One command plus a key — npx crawleo-mcp, then supply credentials

See installation options

USE CASES

01

Getting a page into the model in a shape it can read

Fetch one URL and take it back as markdown instead of raw markup, or as HTML when the structure is what matters.

crawl_web

02

Seeing the results a user in another country would get

Run the same query with the country, locale and device of the audience you are checking for, across more than one page of results.

search_web

2 tools

Counted once with only startup setup. Other versions, modes or optional credentials may expose a different list.

`crawl_web`

Crawl a specific webpage and extract its content in various formats including Markdown, raw HTML, and AI-enhanced HTML.

BEFORE YOU INSTALL

What to know first

Tools you get

2 tools when we connected. The npm build registers search_web and crawl_web, while the documentation uses web.search and web.crawl; check tools/list on the route you use. Both are read-only.

Requires

npx on PATH
CRAWLEO_API_KEY — a Crawleo API key (starts with sk_) from your crawleo.dev dashboard

CONTEXT COST

Light Standing cost

Rated by the size of the 2 tool definitions that ride along on every turn, not by how many there are

INSTALLATION

Connect your client

Server setup

npx

TERMINAL

npx crawleo-mcp

Hosted endpoint

ENDPOINT URL

https://api.crawleo.dev/mcp

Add this hosted endpoint to your MCP client using the credentials listed under Requirements.

MCP Clients

Cursor

CURSOR · JSON

{
  "mcpServers": {
    "crawleo": {
      "url": "https://api.crawleo.dev/mcp",
      "headers": {
        "Authorization": "Bearer <YOUR_CRAWLEO_API_KEY>"
      }
    }
  }
}

Global: ~/.cursor/mcp.json · Per project: .cursor/mcp.json in the project folder

Windsurf

WINDSURF · JSON

{
  "mcpServers": {
    "crawleo": {
      "serverUrl": "https://api.crawleo.dev/mcp",
      "headers": {
        "Authorization": "Bearer <YOUR_CRAWLEO_API_KEY>"
      }
    }
  }
}

~/.codeium/windsurf/mcp_config.json

GitHub Copilot

GITHUB COPILOT · JSON

{
  "servers": {
    "Crawleo": {
      "type": "http",
      "url": "https://api.crawleo.dev/mcp",
      "headers": {
        "Authorization": "Bearer <YOUR_CRAWLEO_API_KEY>"
      }
    }
  }
}

OpenAI Responses API

OPENAI RESPONSES API · JSON

{
  "type": "mcp",
  "server_label": "Crawleo",
  "server_url": "https://api.crawleo.dev/mcp",
  "server_description": "Crawleo MCP Server - Real-Time Web Knowledge for AI",
  "authorization": "<YOUR_CRAWLEO_API_KEY>",
  "allowed_tools": [
    "web.search",
    "web.crawl"
  ],
  "require_approval": "always"
}

Any MCP client — npx

ANY MCP CLIENT — NPX · JSON

{
  "mcpServers": {
    "crawleo": {
      "command": "npx",
      "args": [
        "crawleo-mcp"
      ],
      "env": {
        "CRAWLEO_API_KEY": "<YOUR_CRAWLEO_API_KEY>"
      }
    }
  }
}

<YOUR_CRAWLEO_API_KEY> is your Crawleo API key (it starts with sk_; copy it from your dashboard at crawleo.dev).

Any MCP client — Docker

Clone the repository and build the image

ANY MCP CLIENT — DOCKER · TERMINAL

git clone https://github.com/Crawleo/Crawleo-MCP.git
cd Crawleo-MCP
docker build -t crawleo-mcp .

Add this to your client's MCP configuration

<YOUR_CRAWLEO_API_KEY> is your sk_ key from the crawleo.dev dashboard.

ANY MCP CLIENT — DOCKER · JSON

{
  "mcpServers": {
    "crawleo": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "-e",
        "CRAWLEO_API_KEY=<YOUR_CRAWLEO_API_KEY>",
        "crawleo-mcp"
      ]
    }
  }
}

Any MCP client — local build

Clone and build the server

ANY MCP CLIENT — LOCAL BUILD · TERMINAL

git clone https://github.com/Crawleo/Crawleo-MCP.git
cd Crawleo-MCP
npm install
npm run build

Add this to your client's MCP configuration

<YOUR_PATH_TO_CRAWLEO_MCP> is the folder you cloned into, and <YOUR_CRAWLEO_API_KEY> is your sk_ key.

ANY MCP CLIENT — LOCAL BUILD · JSON

{
  "mcpServers": {
    "crawleo": {
      "command": "node",
      "args": [
        "<YOUR_PATH_TO_CRAWLEO_MCP>/dist/index.js"
      ],
      "env": {
        "CRAWLEO_API_KEY": "<YOUR_CRAWLEO_API_KEY>"
      }
    }
  }
}

Crawleo MCP Server