> This page is for Agent Handler.

> For clean Markdown of any page, append .md to the page URL.
> For a complete documentation index, see https://docs.merge.dev/llms.txt.
> For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://docs.merge.dev/_mcp/server.

# Firecrawl

**Authentication:** API key. See [Magic Link](/merge-agent-handler/features/magic-link) for the runtime auth flow.

## Sample use cases

* Crawl `acme.com/blog` and extract every post title, author, and publish date.
* Scrape the pricing page from our top 5 competitors and diff against last week.
* Get the structured data from this product listing page.

## Available Tools

#### start\_agent

Start an AI agent that searches, navigates, and extracts data from websites. Provide a prompt describing what to find. Check status with get\_agent\_status.

#### get\_agent\_status

Get status and results of an AI agent job. Returns extracted data when complete, or error if failed.

#### start\_batch\_scrape

Start async batch scrape for multiple URLs. More efficient than individual scrapes. Returns job ID to check with get\_batch\_scrape\_status.

#### get\_batch\_scrape\_status

Get status and results of a batch scrape job. Returns progress and scraped data for completed URLs.

#### start\_crawl

Start an async crawl job to scrape multiple pages from a website. Returns a job ID to check status. Use get\_crawl\_status to monitor progress.

#### get\_crawl\_status

Get status and results of a crawl job. Returns progress (total/completed pages) and scraped data when available.

#### cancel\_crawl

Cancel a running crawl job. Use to stop a crawl that is taking too long or no longer needed.

#### start\_extract

Start AI extraction to pull structured data from URLs. Provide a prompt or JSON schema. Returns job ID to check with get\_extract\_status.

#### get\_extract\_status

Get status and results of an AI extraction job. Returns extracted data matching your prompt or schema when complete.

#### map\_site

Discover all URLs on a website. Returns a list of URLs found via sitemap and crawling. Use to find pages before crawling or scraping.

#### scrape

Scrape a URL and extract content as markdown, HTML, or screenshot. Supports browser actions, mobile mode, proxy, and ad blocking.

#### search

Search the web with a query and optionally scrape results. Returns search results with markdown content. Configure language and country for localized results.