MCP server · Web + search
Firecrawl MCP
Search, scrape, parse and crawl the live web through Firecrawl, with optional browser interaction and autonomous research.
Quick verdict
A powerful flagship web connector whose value is its breadth, provided the profile does not mislabel the full endpoint as read-only or hide its variable-cost and remote-interaction surface.
- MaintainerFirecrawl (official)
- Security scopeRead-write
- Transportsstreamable-http, stdio, sse
- Install channelremote · https://mcp.firecrawl.dev/v2/mcp
- EvidenceSource Validated
- LicenseMIT · Open source
Firecrawl MCP gives assistants a broad hosted web-data surface: search, page scraping, file parsing, mapping, crawling, research and monitoring, plus an interactive browser for content that appears only after user actions. The full endpoint is not purely read-only, so the profile must distinguish retrieval from browser, feedback and monitoring actions.
Install Firecrawl MCP
http https://mcp.firecrawl.dev/v2/mcp
{
"mcpServers": {
"firecrawl": {
"url": "https://mcp.firecrawl.dev/v2/mcp",
"headers": {
"Authorization": "Bearer <FIRECRAWL_API_KEY>"
}
}
}
}Firecrawl-hosted endpoint; a rate-limited keyless subset is available without FIRECRAWL_API_KEY. The open-source firecrawl-mcp-server package can instead run locally over stdio, local Streamable HTTP or SSE, or point at a self-hosted Firecrawl API.
Credentials
| Name | Required | What it is for |
|---|---|---|
FIRECRAWL_API_KEY | Required | Firecrawl API key for the full hosted tool set and higher limits. Store it only in the client's secure header or secret setting. |
Verified configurations
Not run by us yet. This profile is Source Validated: the install channel, config entry, credentials and license were checked against the source and upstream metadata, but no configuration has been installed, discovered and executed in a recorded environment. When one is, it appears here with the eight scope dimensions below.
Best for
Agents that need clean, structured web content across individual pages, whole sites or multi-source research rather than simple URL fetching.
Why use it
It combines first-party search, scraping, crawling, parsing and browser-backed extraction behind one MCP endpoint with built-in rate handling and structured outputs.
Tradeoffs
Broad crawling can consume credits quickly, return untrusted or prompt-injected content, and place load on third-party sites. Interactive tools can click or type on real pages, feedback tools submit data to Firecrawl, and autonomous research widens both cost and behavioral uncertainty. Use tight URL/domain limits, a dedicated API key, budget controls and explicit approval before any interaction tool.
Upstream state
repository active · in the official MCP Registry (active). Lifecycle is derived from a committed upstream snapshot at build time, never authored by hand.
Official MCP Registry name: io.github.firecrawl/firecrawl-mcp-server
Editorial review
Grade B — Source Validated. Re-review is due by 2027-02-18. Config entries, credential names and package status can change upstream; verify against the source repository before granting an agent access.
Sources
Related MCP servers
- GitReference MCP server for reading, searching and changing a local Git repository — status, diffs, log, commits, branches and checkouts.
- Tavily MCPSearch, extract, map and crawl live web content through Tavily's official hosted MCP server.
- Exa MCPSearch the web and fetch clean page content through Exa, with optional advanced search and multi-step research tools.