> ## Documentation Index
> Fetch the complete documentation index at: https://firecrawl-docs-agent-add-spark-2.mintlify.site/llms.txt
> Use this file to discover all available pages before exploring further.

# MCP Web Search & Scrape in Claude Code

> Add web scraping and search to Claude Code in 2 minutes

Connect Claude Code to Firecrawl through the hosted MCP server. OAuth is the recommended interactive setup because the key never enters the command line, project files, or conversation.

## Quick setup

### 1. Add Firecrawl

Run:

```bash theme={null}
claude mcp add --transport http firecrawl https://mcp.firecrawl.dev/v2/mcp-oauth
```

`https://mcp.firecrawl.dev/v2/mcp-oauth` is a client configuration value, not a page to open directly in a browser.

### 2. Sign in and verify

Enter `/mcp` in Claude Code, choose `firecrawl`, and complete the browser sign-in. Start a new Claude Code session if you changed an existing Firecrawl connection.

Confirm that only one `firecrawl` server is active, then try:

```text theme={null}
Search for the latest Next.js release notes and summarize the sources.
```

```text theme={null}
Scrape https://firecrawl.dev and explain what it does.
```

## Other connection modes

* [Try keyless MCP](/mcp-server/keyless#try-keyless) for Search, Scrape, and Parse without an account or key.
* [Configure an API key](/mcp-server/keyless#add-an-api-key) for unattended use. Keep the secret outside agent chat and use an environment-variable reference.
* [Run MCP locally](/mcp-server/local) only when you need a local process or a self-hosted Firecrawl API. The local package requires Node.js 22 or newer.

## Optional: literature research

For paper and literature work, the research-index skill routes those queries to the `firecrawl_research_*` tools and the [Research Index](/features/research) instead of general web search:

```bash theme={null}
npx skills add firecrawl/skills@firecrawl-research-index
```

## Troubleshooting

* **A Firecrawl server already exists:** run `claude mcp remove firecrawl`, add the intended connection once, then start a new session.
* **The browser does not open:** enter `/mcp`, select `firecrawl`, and start authentication from the client.
* **The server shows `401`:** sign in again through `/mcp`; do not paste an API key into the conversation or OAuth URL.