MCP Server

Scrapering exposes an MCP (Model Context Protocol) server over HTTP. Connect any compatible AI agent or IDE to give it the ability to scrape URLs on demand.

Requires the url-scraper scope on your API key.

Endpoint

POST https://app.scrapering.com/api/mcp

Available Tools

parse_url

Fetches full HTML content of a URL using a real browser. Returns content, cookies, and optional screenshot. Mirrors the URL Scraper API.

Connect to Claude Desktop

Edit ~/Library/Application Support/Claude/claude_desktop_config.json (macOS) or %APPDATA%\Claude\claude_desktop_config.json (Windows):

{
  "mcpServers": {
    "scrapering": {
      "url": "https://app.scrapering.com/api/mcp",
      "headers": {
        "Authorization": "Bearer YOUR_API_KEY"
      }
    }
  }
}

Restart Claude Desktop. The parse_url tool will appear in the tools list.

Connect to Cursor

Create or edit .cursor/mcp.json in your project root (project-scoped) or ~/.cursor/mcp.json (global):

{
  "mcpServers": {
    "scrapering": {
      "url": "https://app.scrapering.com/api/mcp",
      "headers": {
        "Authorization": "Bearer YOUR_API_KEY"
      }
    }
  }
}

Open Cursor Settings → MCP to verify the server shows as connected.

Connect to Windsurf

Edit ~/.codeium/windsurf/mcp_config.json:

{
  "mcpServers": {
    "scrapering": {
      "url": "https://app.scrapering.com/api/mcp",
      "headers": {
        "Authorization": "Bearer YOUR_API_KEY"
      }
    }
  }
}

Reload Windsurf. The tool will be available in Cascade.

Connect to VS Code (GitHub Copilot)

Create .vscode/mcp.json in your workspace:

{
  "servers": {
    "scrapering": {
      "type": "http",
      "url": "https://app.scrapering.com/api/mcp",
      "headers": {
        "Authorization": "Bearer YOUR_API_KEY"
      }
    }
  }
}

Enable Chat: MCP in VS Code settings if not already active.

Connect to Claude Code

Add via CLI (writes to ~/.claude.json by default):

claude mcp add --transport http scrapering https://app.scrapering.com/api/mcp \
  --header "Authorization: Bearer YOUR_API_KEY"

Or add to .mcp.json in your project root to share with your team:

{
  "mcpServers": {
    "scrapering": {
      "type": "http",
      "url": "https://app.scrapering.com/api/mcp",
      "headers": {
        "Authorization": "Bearer YOUR_API_KEY"
      }
    }
  }
}

Connect to Codex CLI

Edit ~/.codex/config.toml:

[mcp_servers.scrapering]
url = "https://app.scrapering.com/api/mcp"
bearer_token_env_var = "SCRAPERING_API_KEY"
enabled = true

bearer_token_env_var is the name of the environment variable holding your key, not the key itself. Set it before running:

export SCRAPERING_API_KEY=YOUR_API_KEY

Or add via CLI:

codex mcp add scrapering \
  --url https://app.scrapering.com/api/mcp \
  --bearer-token-env-var SCRAPERING_API_KEY

Connect to opencode

Edit ~/.config/opencode/opencode.json (global) or ./opencode.json (project):

{
  "mcp": {
    "scrapering": {
      "type": "remote",
      "url": "https://app.scrapering.com/api/mcp",
      "headers": {
        "Authorization": "Bearer YOUR_API_KEY"
      }
    }
  }
}

Connect to OpenClaw

Edit ~/.openclaw/openclaw.json:

{
  "mcp": {
    "servers": {
      "scrapering": {
        "url": "https://app.scrapering.com/api/mcp",
        "transport": "streamable-http",
        "headers": {
          "Authorization": "Bearer YOUR_API_KEY"
        }
      }
    }
  }
}

Or add via CLI:

openclaw mcp add scrapering \
  --url https://app.scrapering.com/api/mcp \
  --transport streamable-http

Usage Examples

Once connected, prompt your AI agent naturally:

Scrape https://example.com and summarize the main content.

Fetch https://example.com on mobile and tell me what the page title is.

Scrape https://example.com, take a screenshot, and describe the layout.

The agent will call parse_url automatically and work with the returned HTML and cookies.

Next steps

• Authentication — manage API keys and set the url-scraper scope
• URL Scraper — use the REST API directly instead of MCP
• Pricing — each parse_url call costs the same as a URL Scraper request