# Firecrawl MCP Server

> Lets your agent search, scrape, crawl and interact with the live web, and get pages back as clean Markdown or structured data.

Firecrawl's own MCP server brings Firecrawl, its context API for AI agents, into your agent. It searches the web, turns any page into clean Markdown or structured JSON, maps and crawls sites, works through pages in a real browser and runs research jobs across several sites.

What you can do with it
- Search the web, with the full content of the results when you need it
- Scrape any URL into Markdown, or into JSON that matches a schema you give it
- Map a site to find its URLs, or crawl a section for content from many pages
- Parse PDFs, Word documents, spreadsheets and other files
- Click, type, fill forms and navigate dynamic pages in a real browser
- Run Agent jobs that research across several sites and return structured data
- Monitor pages on a schedule and get told when their content changes
- Search developer sources (GitHub issues, pull requests, READMEs and docs), US government sources, and scientific papers from PubMed, bioRxiv, medRxiv and arXiv
- Check your Firecrawl credit usage

How it connects
Firecrawl hosts it at https://mcp.firecrawl.dev/v2/mcp. With no key, the keyless free tier lets you scrape, search and parse within daily limits. Add your Firecrawl API key as an Authorization: Bearer header to unlock every tool and your plan's limits, or point your agent at https://mcp.firecrawl.dev/v2/mcp-oauth to sign in to Firecrawl in your browser instead. Firecrawl also offers native connectors for ChatGPT, Codex and Claude.

To run it yourself, use npx -y firecrawl-mcp with FIRECRAWL_API_KEY set. It also comes as a Docker image and a Gemini CLI extension, and it can use a self-hosted Firecrawl through FIRECRAWL_API_URL.

Good to know
- Set your API key in your agent's settings; never put it in the server address or a chat.
- Crawl, map and Agent jobs need a key or sign-in; they aren't part of the keyless tier.
- Calls use your team's Firecrawl credits.
- A search-only address, https://mcp.firecrawl.dev/v2/mcp-search, offers just the search tools.

- Type: Plugin
- Made by: firecrawl
- Category: Web & browser
- Formats: Remote MCP, Local MCP, Extensions
- Version: 3.30.1
- Website: https://docs.firecrawl.dev/mcp-server
- Source: https://github.com/firecrawl/firecrawl-mcp-server
- Page: https://pluginboard.ai/plugins/firecrawl-mcp

## Works with

- ChatGPT
- Claude
- Muse
- Grok Bot
- Gemini
- Grok
- GitHub Copilot
- Cursor
- Claude Code
- Codex
- Antigravity
- Gemini CLI
- Replit
- OpenClaw
- OpenCode
- Hermes Agent

## Install

You'll need:

- `Authorization` (optional): Optional. Bearer <FIRECRAWL_API_KEY>. Get a key at https://firecrawl.dev/app/api-keys

Replace each `<NAME>` placeholder with your own value.

### ChatGPT

In ChatGPT on the web: Plugins → + → Add custom MCP server, then paste https://mcp.firecrawl.dev/v2/mcp under Connection. Business and Enterprise plans; Pro can add read-only servers.

### Claude

In Claude (web or desktop): Customize → Connectors → Add custom connector, then paste https://mcp.firecrawl.dev/v2/mcp

### Muse

In Muse: In a Muse chat, ask it to create a custom connector for the MCP server at https://mcp.firecrawl.dev/v2/mcp. Then tap Connect on the card it shows and sign in, or add your key.

### Grok Bot

In Grok Bot: In a chat with your Bot, say "Add this MCP server: https://mcp.firecrawl.dev/v2/mcp". It confirms the details with you and sets it up, and the tools are ready from your next message.

### Gemini

In Gemini: At gemini.google.com: Settings → Connected Apps → Add a custom app, then paste https://mcp.firecrawl.dev/v2/mcp

### Grok

In Grok on the web: At grok.com/connectors: New Connector → Custom, then enter https://mcp.firecrawl.dev/v2/mcp and sign in if it asks.

### GitHub Copilot

Add to VS Code: vscode:mcp/install?%7B%22name%22%3A%22firecrawl-mcp-server%22%2C%22type%22%3A%22http%22%2C%22url%22%3A%22https%3A%2F%2Fmcp.firecrawl.dev%2Fv2%2Fmcp%22%7D

Copilot CLI:

```shell
copilot mcp add --transport http firecrawl-mcp-server https://mcp.firecrawl.dev/v2/mcp
```

### Cursor

Add to Cursor: cursor://anysphere.cursor-deeplink/mcp/install?name=firecrawl-mcp-server&config=eyJ1cmwiOiJodHRwczovL21jcC5maXJlY3Jhd2wuZGV2L3YyL21jcCJ9

Add to .cursor/mcp.json:

```json
{
  "mcpServers": {
    "firecrawl-mcp-server": {
      "url": "https://mcp.firecrawl.dev/v2/mcp"
    }
  }
}
```

### Claude Code

Run in your terminal:

```shell
claude mcp add --transport http firecrawl-mcp-server https://mcp.firecrawl.dev/v2/mcp
```

### Codex

Run in your terminal:

```shell
codex mcp add firecrawl-mcp-server --url https://mcp.firecrawl.dev/v2/mcp
```

### Antigravity

Add to ~/.gemini/config/mcp_config.json:

```json
{
  "mcpServers": {
    "firecrawl-mcp-server": {
      "url": "https://mcp.firecrawl.dev/v2/mcp"
    }
  }
}
```

### Gemini CLI

Run in your terminal:

```shell
gemini extensions install https://github.com/firecrawl/firecrawl-mcp-server
```

### Replit

In Replit: At replit.com/integrations, open MCP Servers → + Add MCP server, paste https://mcp.firecrawl.dev/v2/mcp (add any headers under Advanced settings), then click Test & save.

### OpenClaw

Run in your terminal:

```shell
openclaw mcp add firecrawl-mcp-server --url https://mcp.firecrawl.dev/v2/mcp --transport streamable-http
```

If it needs a key or sign-in: For https://mcp.firecrawl.dev/v2/mcp, add --header NAME=value for each header it needs, or --auth oauth and then run openclaw mcp login firecrawl-mcp-server.

### OpenCode

Run in your terminal:

```shell
opencode mcp add firecrawl-mcp-server --url https://mcp.firecrawl.dev/v2/mcp
```

If it needs a key: For https://mcp.firecrawl.dev/v2/mcp, add --header 'NAME=value' for each header it needs. Add --global to use it in every project.

### Hermes Agent

Run in your terminal:

```shell
hermes mcp add firecrawl-mcp-server --url https://mcp.firecrawl.dev/v2/mcp
```

Add to ~/.hermes/config.yaml:

```json
{
  "mcp_servers": {
    "firecrawl-mcp-server": {
      "url": "https://mcp.firecrawl.dev/v2/mcp"
    }
  }
}
```

If it signs in with OAuth: If https://mcp.firecrawl.dev/v2/mcp asks you to sign in, add --auth oauth to the command, or auth: oauth to its entry.

---
From Pluginboard (https://pluginboard.ai).