# Oxylabs

> Scrape any URL, render JavaScript, and reach geo-restricted data — official Oxylabs Web Scraper MCP.

[Canonical HTML page](https://top-mcps.com/mcp/oxylabs) · [server.json](https://top-mcps.com/mcp/oxylabs.json) · [methodology](https://top-mcps.com/about/methodology)

## Install

### Claude Desktop — `claude_desktop_config.json`

Paste under mcpServers. Fully quit and reopen Claude after editing.

```json
{
  "mcpServers": {
    "oxylabs": {
      "command": "uvx",
      "args": [
        "oxylabs-mcp"
      ],
      "env": {
        "OXYLABS_USERNAME": "${OXYLABS_USERNAME}",
        "OXYLABS_PASSWORD": "${OXYLABS_PASSWORD}"
      }
    }
  }
}
```

### Claude Code — `CLI or .mcp.json`

Run from your repo. Commit .mcp.json to share with your team.

```shell
# export OXYLABS_USERNAME=your_username
# export OXYLABS_PASSWORD=your_password
claude mcp add oxylabs -- uvx oxylabs-mcp
```

### Cursor — `.cursor/mcp.json`

Global path: ~/.cursor/mcp.json. Reload window after editing.

```json
{
  "mcpServers": {
    "oxylabs": {
      "command": "uvx",
      "args": [
        "oxylabs-mcp"
      ],
      "env": {
        "OXYLABS_USERNAME": "${OXYLABS_USERNAME}",
        "OXYLABS_PASSWORD": "${OXYLABS_PASSWORD}"
      }
    }
  }
}
```

### VS Code — `.vscode/mcp.json`

VS Code uses the "servers" key (not "mcpServers").

```jsonc
{
  "servers": {
    "oxylabs": {
      "command": "uvx",
      "args": [
        "oxylabs-mcp"
      ],
      "env": {
        "OXYLABS_USERNAME": "${OXYLABS_USERNAME}",
        "OXYLABS_PASSWORD": "${OXYLABS_PASSWORD}"
      }
    }
  }
}
```

### Windsurf — `~/.codeium/windsurf/mcp_config.json`

Open via Cascade → hammer icon → Configure.

```json
{
  "mcpServers": {
    "oxylabs": {
      "command": "uvx",
      "args": [
        "oxylabs-mcp"
      ],
      "env": {
        "OXYLABS_USERNAME": "${OXYLABS_USERNAME}",
        "OXYLABS_PASSWORD": "${OXYLABS_PASSWORD}"
      }
    }
  }
}
```

### Cline — `cline_mcp_settings.json`

Open via the Cline sidebar → MCP Servers → Edit.

```json
{
  "mcpServers": {
    "oxylabs": {
      "command": "uvx",
      "args": [
        "oxylabs-mcp"
      ],
      "env": {
        "OXYLABS_USERNAME": "${OXYLABS_USERNAME}",
        "OXYLABS_PASSWORD": "${OXYLABS_PASSWORD}"
      }
    }
  }
}
```

### Continue — `~/.continue/config.json`

Continue uses modelContextProtocolServers with a transport block.

```json
{
  "experimental": {
    "modelContextProtocolServers": [
      {
        "transport": {
          "type": "stdio",
          "command": "uvx",
          "args": [
            "oxylabs-mcp"
          ],
          "env": {
            "OXYLABS_USERNAME": "${OXYLABS_USERNAME}",
            "OXYLABS_PASSWORD": "${OXYLABS_PASSWORD}"
          }
        }
      }
    ]
  }
}
```

### Codex CLI — `~/.codex/config.toml`

Codex uses TOML. Each server is a [mcp_servers.<name>] subtable.

```shell
# ~/.codex/config.toml
[mcp_servers.oxylabs]
command = "uvx"
args = [
  "oxylabs-mcp",
]
env = { OXYLABS_USERNAME = "${OXYLABS_USERNAME}", OXYLABS_PASSWORD = "${OXYLABS_PASSWORD}" }
```

### Zed — `~/.config/zed/settings.json`

Zed calls them "context_servers". Settings live-reload on save.

```jsonc
{
  "context_servers": {
    "oxylabs": {
      "command": {
        "path": "uvx",
        "args": [
          "oxylabs-mcp"
        ]
      },
      "env": {
        "OXYLABS_USERNAME": "${OXYLABS_USERNAME}",
        "OXYLABS_PASSWORD": "${OXYLABS_PASSWORD}"
      }
    }
  }
}
```

### ChatGPT — `ChatGPT → Apps directory`

Oxylabs doesn't ship a hosted HTTPS endpoint today. ChatGPT supports remote MCP servers only — to use this server in ChatGPT you'll need to deploy it to a public HTTPS URL first (e.g. via Cloudflare Workers or Vercel) or wait for an official remote build.

```none

```

## At a glance

- **Maintainer:** Oxylabs
- **Transport:** stdio
- **Auth model:** API key
- **Required secrets:** OXYLABS_USERNAME, OXYLABS_PASSWORD
- **Supported clients:** Claude, Cursor, VS Code, Any MCP-compatible client, Oxylabs accounts
- **License:** MIT
- **Language:** Python
- **Latest version:** latest
- **Last verified:** 2026-07-16
- **Score:** 67/100 (rubric 2026-04 — see https://top-mcps.com/about/methodology)
- **Source:** https://github.com/oxylabs/oxylabs-mcp

## Quick answer

**What it does.** Exposes the Oxylabs Web Scraper API as MCP tools: universal_scraper, google_search_scraper, amazon_search/product_scraper, plus optional AI Studio scraper, crawler, and browser-agent tools.

**Best for:**
- Structured Google + Amazon scraping
- JavaScript-rendered page extraction
- Geo-restricted data (195+ countries)
- High-volume SERP + e-commerce data
- AI Studio crawler + browser-agent tools

**Not for:**
- Sites with an existing API
- Low-volume fetches Fetch/Firecrawl handle
- Teams that cannot use a paid data network

## Description

Oxylabs' official MCP (MIT, Python) bridges an agent to the Oxylabs Web Scraper API: a universal scraper for any URL, dedicated Google and Amazon parsers, JavaScript rendering, CAPTCHA handling, and geo-routing across 195+ countries. With an AI Studio key it also exposes AI scraper, crawler, and browser-agent tools.

## Why it matters

E-commerce and SERP scraping at scale needs parsed, structured output and reliable geo-routing. Oxylabs is one of the two enterprise data networks (with Bright Data) purpose-built for that.

## Key features

- Universal scraper + Google/Amazon parsers
- JS rendering + CAPTCHA handling
- Geo-routing across 195+ countries
- Optional AI Studio tools (scraper, crawler, browser agent)
- Official Oxylabs, MIT, Python (uvx)

## FAQ

### Which tools show up?

With OXYLABS_USERNAME/PASSWORD you get the Web Scraper API tools (universal, Google, Amazon). Add an OXYLABS_AI_STUDIO_API_KEY to also expose AI scraper, crawler, and browser-agent tools.

### Bright Data or Oxylabs?

Both are enterprise scraping networks. Oxylabs leans into parsed SERP and e-commerce (Google, Amazon) endpoints; Bright Data leads on the Web Unlocker for arbitrary hostile targets. Pick by whether you need structured verticals or raw unblocking.

### How do I run it?

It is a Python package — `uvx oxylabs-mcp` runs it with no manual install once uv is present.

## Changelog

- **2026-07-16** — Refreshed install snippets and fact sheet; verified for 2026.
- **2025-05-01** — Initial directory listing.