# CrawlEyes MCP server

> Enables AI agents to perform web scraping, semantic search, and deep research using SearXNG, Tavily, and Crawl4AI.

- Repository: https://github.com/waiky-github/CrawlEyes
- Page: https://hostedmcp.org/servers/waiky-github/crawleyes
- Transport: stdio
- Stars: 2
- License: MIT
- Latest version: 0.2.0

## Install

### Claude Code

```sh
claude mcp add --env 'TAVILY_API_KEY=<YOUR_TAVILY_API_KEY>' crawleyes -- uvx crawleyes
```

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

```json
{
  "mcpServers": {
    "crawleyes": {
      "command": "uvx",
      "args": [
        "crawleyes"
      ],
      "env": {
        "TAVILY_API_KEY": "<YOUR_TAVILY_API_KEY>"
      }
    }
  }
}
```

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

```json
{
  "servers": {
    "crawleyes": {
      "type": "stdio",
      "command": "uvx",
      "args": [
        "crawleyes"
      ],
      "env": {
        "TAVILY_API_KEY": "<YOUR_TAVILY_API_KEY>"
      }
    }
  }
}
```

### VS Code CLI

```sh
code --add-mcp '{"name":"crawleyes","type":"stdio","command":"uvx","args":["crawleyes"],"env":{"TAVILY_API_KEY":"<YOUR_TAVILY_API_KEY>"}}'
```

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

```json
{
  "mcpServers": {
    "crawleyes": {
      "command": "uvx",
      "args": [
        "crawleyes"
      ],
      "env": {
        "TAVILY_API_KEY": "<YOUR_TAVILY_API_KEY>"
      }
    }
  }
}
```

### Codex

```sh
codex mcp add crawleyes --env 'TAVILY_API_KEY=<YOUR_TAVILY_API_KEY>' -- uvx crawleyes
```

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

```toml
[mcp_servers.crawleyes]
command = "uvx"
args = ["crawleyes"]

[mcp_servers.crawleyes.env]
TAVILY_API_KEY = "<YOUR_TAVILY_API_KEY>"
```

### Gemini CLI

```sh
gemini mcp add -s user -e 'TAVILY_API_KEY=<YOUR_TAVILY_API_KEY>' crawleyes uvx -- crawleyes
```

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

```json
{
  "mcpServers": {
    "crawleyes": {
      "command": "uvx",
      "args": [
        "crawleyes"
      ],
      "env": {
        "TAVILY_API_KEY": "<YOUR_TAVILY_API_KEY>"
      }
    }
  }
}
```

## Environment variables

| Name | Required | Secret | Description |
|---|---|---|---|
| `TAVILY_API_KEY` | no | yes | Optional. Tavily API key for higher-rate search fallback. If unset, uses Tavily keyless (rate-limited). |
| `PLAYWRIGHT_DOWNLOAD_HOST` | no | no |  |
| `SEARXNG_URL` | no | no | Optional. Base URL of a SearXNG instance (e.g. http://localhost:8080). If unset, falls back to Tavily keyless. |

## FAQ

### Does CrawlEyes require an API key?

No, it is designed to work without external accounts or API keys using SearXNG and a keyless Tavily fallback.

### What tools are exposed by the MCP server?

The MCP server exposes search, extract, deep_research, and sitemap tools.

### Can I use CrawlEyes for deep research?

Yes, the deep_research tool decomposes a topic into sub-questions, searches, extracts, and synthesizes a cited Markdown report.

### How does the server handle search failures?

It uses a SearXNG primary search with an automatic fallback to the Tavily keyless API, managed by a three-state circuit breaker.