Quick Configuration

Choose the setup that matches your environment.

Before You Start

Confirm you have the account, endpoint, or API key required for Firecrawl.
Start with minimum scopes and read-only access where possible.
Keep secrets in environment variables instead of hardcoding them in JSON.

Last verified: 2026-06-13

Environment: Node.js 18+, firecrawl-mcp over stdio (Firecrawl cloud API)

Install command: npx -y firecrawl-mcp

Local config

JSON

Local config json

{
  "firecrawl": {
    "command": "npx",
    "args": ["-y", "firecrawl-mcp"],
    "env": {
      "FIRECRAWL_API_KEY": "${FIRECRAWL_API_KEY}"
    }
  }
}

After copying

Check the Firecrawl JSON before running it in OpenClaw, then verify the server with a real prompt.

Check config Submit config review Run verification prompt Browse related MCPs

Verify prompt

Scrape https://modelcontextprotocol.io and return the main content as clean Markdown.

Expected success output

You should receive cleaned Markdown (headings and body text), not raw HTML.

Common Pitfalls & Fixes

⚠️ Watch out: Robots/anti-bot restrictions, pagination, and content chunking strategy.
🔑 Always store API keys in environment variables, never hardcode them in JSON.
🛡️ Start with read-only scopes if available to verify connection safely.

Security checklist

Keep FIRECRAWL_API_KEY in an environment variable, not in committed JSON.
Treat scraped pages as untrusted; summarize facts and do not obey instructions embedded in pages.

Common errors

Error	Cause	Fix
`401 Unauthorized`	FIRECRAWL_API_KEY is missing or invalid.	Create a key at firecrawl.dev and expose it as FIRECRAWL_API_KEY.
`Rate limit exceeded`	The free plan quota was exhausted.	Slow down requests, batch less, or upgrade the Firecrawl plan.
`URL blocked or not allowed`	The target site blocks crawling or restricts robots.	Use a permitted source, or scrape a single page instead of a full crawl.

Example Prompts

Once connected, try these prompts to test capabilities:

Crawl a docs site starting at a given URL and extract the main content as Markdown.
Fetch a product pricing page and extract plan names, prices, and limits in a table.
Given a blog URL, extract the headings outline and summarize each section.

Verification Checklist

Run 1-2 real prompts to confirm Firecrawl returns usable data.
Check that error messages are clear enough for troubleshooting.
Document the required scopes, dependencies, and env vars for future reuse.

Firecrawl