# fastCRW MCP server

> Converts URLs into clean markdown or structured JSON through web scraping, crawling, mapping, and searching capabilities.

- Repository: https://github.com/us/crw
- Page: https://hostedmcp.org/servers/us/crw
- Transport: stdio
- Stars: 1260
- License: AGPL-3.0
- Latest version: 0.36.0

## Install

### Claude Code

```sh
claude mcp add crw -- docker run -i --rm ghcr.io/us/crw:latest
```

### Cursor (`~/.cursor/mcp.json`)

```json
{
  "mcpServers": {
    "crw": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "ghcr.io/us/crw:latest"
      ]
    }
  }
}
```

### VS Code (`.vscode/mcp.json`)

```json
{
  "servers": {
    "crw": {
      "type": "stdio",
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "ghcr.io/us/crw:latest"
      ]
    }
  }
}
```

### VS Code CLI

```sh
code --add-mcp '{"name":"crw","type":"stdio","command":"docker","args":["run","-i","--rm","ghcr.io/us/crw:latest"]}'
```

### Claude Desktop (`claude_desktop_config.json`)

```json
{
  "mcpServers": {
    "crw": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "ghcr.io/us/crw:latest"
      ]
    }
  }
}
```

### Codex

```sh
codex mcp add crw -- docker run -i --rm ghcr.io/us/crw:latest
```

### Codex config.toml (`~/.codex/config.toml`)

```toml
[mcp_servers.crw]
command = "docker"
args = ["run", "-i", "--rm", "ghcr.io/us/crw:latest"]
```

### Gemini CLI

```sh
gemini mcp add -s user crw docker -- run -i --rm ghcr.io/us/crw:latest
```

### Gemini settings.json (`~/.gemini/settings.json`)

```json
{
  "mcpServers": {
    "crw": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "ghcr.io/us/crw:latest"
      ]
    }
  }
}
```

## Environment variables

| Name | Required | Secret | Description |
|---|---|---|---|
| `CRW_API_KEY` | no | no |  |

## Tools (9)

- `crw_scrape`: Scrape one URL to Markdown, HTML, plain text, links, or images. (read-only)
- `crw_crawl`: Start an async site crawl; returns a job id to poll with crw_check_crawl_status.
- `crw_check_crawl_status`: Poll an async crawl job and retrieve its pages. (read-only)
- `crw_map`: Discover URLs on a site via sitemap and/or a short crawl. Returns a URL list only, no page content. (read-only)
- `crw_extract`: Extract structured JSON from URLs via a prompt and/or JSON schema. Async job — poll crw_check_extract_status with the returned id. Needs an LLM.
- `crw_check_extract_status`: Poll an extract job; returns status and, when complete, a per-URL results array. (read-only)
- `crw_cancel_extract`: Request cancellation of an extract job. Returns the canonical status; cancelling remains non-terminal until the claimed URL settles.
- `crw_parse_file`: Parse a local PDF (base64 in contentBase64) to markdown. No OCR: scanned PDFs return empty markdown with a warning. (read-only)
- `crw_enrich`: Fill blank cells of up to 3 rows from the web. Reads each row's website, searches for the rest; every cell carries a source url, excerpt and status. Trust status 'supported'; 'notFound' = no page stated it. Needs an LLM. (read-only)

## FAQ

### How do I install it?

You can install it using a single curl command, which runs locally for free or connects to the managed API if you provide an API key.

### Does it support AI coding tools?

Yes, it automatically detects and integrates with tools like Claude Code, Cursor, Codex, Gemini CLI, OpenCode, and Windsurf.

### Can I run it myself?

Yes, you can run it locally as a small Rust binary for data control and private networks, or use the managed API for zero infrastructure.
