How to use Firecrawl with Cursor
You want Firecrawl inside Cursor so it can research a technical question across documentation, extract structured records from a website, audit a website before an agent uses it, and investigate issues and research papers — without you copy-pasting between tabs. Here is the whole setup, once.
Create your Firecrawl gateway
The gateway is the thing Cursor connects to, and it is where you decide which Firecrawl tools an agent may use. You create it once and every client reuses it.
-
Create an Arcade account
Sign up and open the dashboard. The free tier is enough to follow this through.
-
Create an MCP Gateway and select your Firecrawl tools
In the dashboard, create a gateway and pick the Firecrawl tools you want to expose. The gateway is what your client connects to, and it is where tool access is decided.
-
Copy your gateway URL
Every client below needs it.
https://api.arcade.dev/mcp/<YOUR-GATEWAY-SLUG>
Connect the gateway to Cursor
Cursor does not refresh MCP OAuth tokens automatically. For a persistent connection, set Authorization to “Arcade Headers” on the gateway in the dashboard.
-
Open MCP settings
Command Palette (⌘⇧P / Ctrl+Shift+P) → Open MCP Settings, then click “New MCP Server”.
-
Add the gateway to mcpServers
{ "mcpServers": { "mcp-arcade": { "url": "https://api.arcade.dev/mcp/<YOUR-GATEWAY-SLUG>" } } } -
Try it out
Ask Cursor to use one of your Firecrawl tools.
Full Cursor guide in the docs steps verified 2026-08-13
Try these web research prompts
Cursor can now act on your Firecrawl account. Click a prompt to copy it.
Set policy on CrawlWebsite, ExtractData, and CancelCrawl, and require approval before an agent starts a broad crawl or stops one in progress. Pair it with Contextual Access policies so untrusted page content cannot trigger those actions without a policy check.