firecrawl_agent
Autonomous web research agent that returns structured data when you do not know the URLs or the answer spans several sites. Describe the fields you need, optionally pass a JSON schema and seed URLs, and the agent searches, navigates, reads pages, and returns JSON assembled across sources. Use it for
How to use it
firecrawl_agent is exposed by the Firecrawl MCP Server MCP server. Add the server to your MCP client (Claude Desktop, Cursor, Windsurf and others), and the firecrawl_agent tool becomes available to the model automatically. See the full listing for setup details and every tool this server provides.
Install Firecrawl MCP Server
npx -y @smithery/cli install @mendableai/mcp-server-firecrawl --client claudeOther tools in Firecrawl MCP Server (34)
Multi-source research, unknown or many sites
Multi-page extraction (with limits)
Programming questions over developer sources
Check the status of an agent job and retrieve results when complete. Use this to poll for results after starting an agent.
Check the status and results of an existing crawl job by ID.
Starts a crawl job, polls until it reaches a terminal state, and returns the final crawl status/data.
Search an index built for coding agents. The index covers GitHub issues, merged pull requests, repository READMEs, and curated documentation sites.
Sends structured feedback for a completed v2 endpoint job through /v2/feedback.
Interact with a fresh URL or with a page that was already opened by firecrawlscrape.
Stop an interact session for a scraped page when you are done interacting.
Map a website to discover all indexed URLs on the site.
get page-level results, including diff, snapshot, judgment.meaningful, and judgment.meaningfulChanges.
list checks, optionally filtered by status.
delete a monitor (destructive; only call when the user intends to remove it).
get one monitor.
list monitors.
trigger a check now.
update fields including goal, judgeEnabled, webhook, and notification.
Parse local files or hosted upload references with Firecrawl's /v2/parse endpoint.
retrieve canonical metadata for one paper ID (arXiv, PMC, PMID, or DOI).
read full-text passages from a specific paper.
expand from one or more anchor papers through the citation graph.
search indexed public GitHub issue, pull-request, and README content.
search paper metadata and abstracts with a natural-language query, with optional author, category, and date filters.
Scrape content from a single URL with advanced options.
Search the web and optionally extract content from search results.
Sends structured feedback on a previous firecrawlsearch result. The first feedback per search id refunds 1 credit and improves Firecrawl's search quality. Idempotent per search id.
Interact with a URL or scraped page
Discovering URLs on a site
Recurring page checks
Files and hosted upload refs
Paper and GitHub repository research
Single page content
Web search for info