Firecrawl

Overview

Firecrawl is a web data API that turns pages into clean markdown or structured data for AI agents and apps. Endpoints cover scrape, crawl, map, search, interact and monitor, all billed from one credit balance. A basic scrape costs 1 credit per page, search costs 2 credits per 10 results and browser interaction costs 2 credits per minute. A hosted MCP server lets Claude, Cursor and similar clients call it directly.

On October 28, 2026 Firecrawl launched Directory, a recruiting collection inside Alexandria, its library of data providers and search indexes. Directory holds more than 240 million professional profiles with employment histories and work contact details, sourced through providers including Apollo, FullEnrich and Data Legion. An agent can search people by job title, company, location and keywords, then pull a person's current role, previous employers, background and work contact details.

Directory runs inside ChatGPT and Claude through the Firecrawl plugin, and in terminal agents through the Firecrawl CLI skills. Recruiters can hand the agent a job description and have it build a shortlist, then use Firecrawl's web search to find a candidate's talks, writing and open source work. Searches are free; credits are used when retrieving detailed profiles and contact information.

Teams pick Firecrawl for one credit meter across people search, web search and scraping. Free gives 1,000 credits a month with no card. Hobby is $19/mo, or $16/mo yearly, for 5,000 credits. Standard is $83/mo yearly for 100,000 credits, Growth is $333/mo yearly, and Scale is $599/mo yearly for 1 million credits. Paid plans can add credits in $5 increments, and a scrape that returns nothing is not charged.

Who it's for

Key features

Directory people search. Search 240M+ professional profiles by title, company, location and keywords from ChatGPT or Claude; searches are free, detailed profiles and work contact details use credits.

Crawl API. Programmatic crawl of pages you need for sourcing or research pipelines.

Structured extract. Clean markdown/JSON out instead of brittle DIY scrapers.

MCP server

https://mcp.firecrawl.dev/v2/mcp Copy URL

First-party MCP at mcp.firecrawl.dev. Official vendor-hosted server. Connect Claude or ChatGPT with your own login. The Firecrawl plugin for ChatGPT and the Claude connector also give your agent Directory people search.

Setup steps and live status · MCP Directory · Check your stack in the MCP readiness checker

Pricing

Checked 4 Nov 2026. Vendors change prices without notice.

See current pricing on Firecrawl (opens in a new tab)

Pros and cons

Integrations

ChatGPT Claude Cursor

From the vendor's own site (opens in a new tab).

How Firecrawl compares

See all Firecrawl alternatives, with what each one does differently and a head to head for every pairing.

Tool	Free tier	From	MCP	API	Compare
Firecrawl (this tool)	✓	$16/mo	✓	✓
Datamagnet Live people and company data API with its own MCP from $39/mo, if you only want profiles and not a scraping stack.	–	$39/mo	✓	✓	Compare
Apify Actor store if you want scrapers as apps, not one scrape API.	✓	$19/mo	✓	✓	Compare
Exa Search-first web API if crawl-and-markdown is not the job.	✓	$4/1k searches	✓	✓	Compare
Clay Spreadsheet GTM if the scrape feeds enrichment columns.	✓	$185/mo	✓	✓	Compare

Datamagnet

$39/mo MCP API

Live people and company data API with its own MCP from $39/mo, if you only want profiles and not a scraping stack.

View listing · Compare with Firecrawl

Apify

$19/mo Free tier MCP API

Actor store if you want scrapers as apps, not one scrape API.

View listing · Compare with Firecrawl

Exa

$4/1k searches Free tier MCP API

Search-first web API if crawl-and-markdown is not the job.

View listing · Compare with Firecrawl

Clay

$185/mo Free tier MCP API

Spreadsheet GTM if the scrape feeds enrichment columns.

View listing · Compare with Firecrawl