scrape_urlLABSCO SUMMARY
Pages that fight scrapers, returned as clean markdown.
The knobs are the point: a wait for JavaScript to settle, ad and consent-banner blocking, tag include and exclude lists, a mobile viewport. And a scrape hands back a session ID, so a page that needs a click can be continued rather than re-fetched.
What it is
An MCP server exposing Firecrawl's API for scraping, crawling, mapping, searching, document parsing, browser automation and academic search.
What you get
- A single URL returned as markdown, HTML, raw HTML, links, a screenshot or a summary
- Navigation, headers, footers and ads stripped so only the article body comes back
- Public PDF and DOCX URLs detected and parsed automatically
- A wait after page load for JavaScript-rendered pages, as long as you need, and a mobile viewport when you want one
- Whole sites crawled, site structures mapped, and the web searched
- A live browser session carried between calls, so a scrape can continue into interaction
Requirements
Calls run against Firecrawl's API and are metered in credits; the enhanced stealth proxy tier costs more than the default auto tier. The page-load timeout is yours to set, with plenty of headroom above the default, which matters on slow or heavily scripted pages.
Setup effort
Add the hosted endpoint plus a key — paste it into your client, then supply credentials
USE CASES
01
Reading a page the model cannot fetch itself
Scrape the URL to markdown with navigation and footers stripped out.
02
Getting content off a page that renders in JavaScript
Set a wait after load so the content exists before capture.
03
Pulling text out of a PDF that lives at a URL
Point the scrape at it — document formats are detected and parsed.
04
Continuing past a click or a consent wall
Carry the live session from the scrape into the browser interaction instead of starting over.
05
Capturing what a page looks like, not just what it says
Request the screenshot format alongside the markdown.
19 tools
Listed in project docs. Your client may show a different list.
batch_scrape_urlsget_batch_scrape_statuscancel_batch_scrapecrawl_urlget_crawl_statuscancel_crawlmap_urlsearch_webparse_documentrun_agentget_agent_statuscancel_agentbrowser_interactbrowser_closesearch_papersget_paperfind_related_paperssearch_githubBEFORE YOU INSTALL
What to know first
Tools you get
MewCP documents 4 meta-tools visible to your client: search, get_schema, list_accounts and call_tool. Its 19 listed Firecrawl tools are found through search and called through call_tool.
Requires
- A MewCP account — the Hobby tier is free (unlimited servers, 20K actions/month)
- The Firecrawl server enabled in your MewCP toolset — the gateway only reaches servers you have enabled
Where it runs
All 3 documented clients connect to a hosted endpoint.
INSTALLATION
Connect your client
Server setup
ENDPOINT URL
https://gateway.mewcp.com/personal/mcpCopy your MewCP Key from the MewCP dashboard → API Keys and use it as <YOUR_MEWCP_KEY> in the client’s Authorization: Bearer request header. Also add Firecrawl to your MewCP toolset and connect your own Firecrawl API key from firecrawl.dev → API Keys inside MewCP; that key stays out of the client configuration.
MCP Clients
VS Code
VS CODE · JSON
{
"servers": {
"mewcp": {
"type": "http",
"url": "https://gateway.mewcp.com/personal/mcp",
"headers": {
"Authorization": "Bearer <YOUR_MEWCP_KEY>"
}
}
}
}Cursor
CURSOR · JSON
{
"mcpServers": {
"mewcp": {
"url": "https://gateway.mewcp.com/personal/mcp",
"headers": {
"Authorization": "Bearer <YOUR_MEWCP_KEY>"
}
}
}
}Claude Desktop
CLAUDE DESKTOP · JSON
{
"mcpServers": {
"mewcp": {
"command": "npx",
"args": [
"-y",
"mcp-remote@latest",
"https://gateway.mewcp.com/personal/mcp",
"--transport",
"http-only",
"--header",
"Authorization: Bearer <YOUR_MEWCP_KEY>"
]
}
}
}
