Web monitoring inside Claude, ChatGPT and Cursor

The PageCrawl MCP server lets Claude, ChatGPT and Cursor set up monitors and read changes for you. Works with any MCP client, and every tool is included on every plan, Free included.

https://mcp.pagecrawl.io/mcp
Connect your AISign up free
Trusted by teams at
MicrosoftNew York TimesDeloitteGoldman SachsHubSpotYahooEYStanford

Add PageCrawl to your AI

Pick your client. You sign in to PageCrawl the first time it connects.

ClaudeOAuth

Customize, then Connectors, then Add custom connector. Paste the server URL.

Manual setup
  1. Open claude.ai or Claude Desktop and go to Customize, then Connectors.
  2. Click + Add, then Add custom connector.
  3. Name it "PageCrawl", set the URL to https://mcp.pagecrawl.io/mcp and click Continue.
  4. Keep the sign-in settings Claude detects, click Continue, then Add. Sign in to PageCrawl and click Allow Claude access.
  5. In a conversation, click + under the message box, open Connectors and turn PageCrawl on.

Note: On Team and Enterprise plans, an Owner adds the connector under Organization settings, then Connectors. Members then click Connect on it.

ChatGPTOAuth

Plugins, then Add custom MCP server. Paste the server URL and choose OAuth.

Manual setup
  1. Open ChatGPT on the web, go to Plugins, select + and choose Add custom MCP server.
  2. Name it "PageCrawl" and set the server URL to https://mcp.pagecrawl.io/mcp.
  3. Choose OAuth, tick I understand and want to continue, and select Create as a plugin.
  4. Sign in to PageCrawl and click Allow ChatGPT access.
  5. Install the new plugin, then in a conversation type @ and pick PageCrawl.

Note: Creating and changing monitors from ChatGPT needs its full MCP support, which OpenAI offers on Business, Enterprise and Edu. On Business only admins and owners can add the server, and on Pro a custom MCP server can only read.

Claude CodeOAuth

Install the plugin: the connection plus skills that teach Claude Code to use it.

/plugin install pagecrawl --marketplace pagecrawl/skills
Manual setup

To connect only the server, add this to .mcp.json in your project root (Claude Code skips an entry without the type).

{
  "mcpServers": {
    "pagecrawl": {
      "type": "http",
      "url": "https://mcp.pagecrawl.io/mcp"
    }
  }
}

Either way, sign in once: run /mcp, choose PageCrawl and approve access in your browser.

CursorOAuth

Cursor opens, adds PageCrawl and asks you to sign in.

Manual setup

Or add this to ~/.cursor/mcp.json:

{
  "mcpServers": {
    "pagecrawl": {
      "url": "https://mcp.pagecrawl.io/mcp"
    }
  }
}

VS CodeOAuth

VS Code opens, adds PageCrawl and asks you to sign in.

Manual setup

Or add this to .vscode/mcp.json:

{
  "servers": {
    "pagecrawl": {
      "type": "http",
      "url": "https://mcp.pagecrawl.io/mcp"
    }
  }
}

CodexOAuth

Run this, then sign in when the browser opens.

codex mcp add pagecrawl --url https://mcp.pagecrawl.io/mcp
Manual setup

If the sign-in does not open, run codex mcp login pagecrawl.

Other clientsAPI token

Cline, Windsurf, OpenClaw and any client that sends custom headers: use an API token.

Manual setup
  1. In PageCrawl, go to Settings, then API access, then Create Token. Copy the token (shown once).
  2. Paste this into your client's MCP config file.
{
  "mcpServers": {
    "pagecrawl": {
      "url": "https://mcp.pagecrawl.io/mcp",
      "headers": {
        "Authorization": "Bearer YOUR_TOKEN_HERE"
      }
    }
  }
}

Cline needs "type": "streamableHttp" beside the URL, or it tries the older SSE transport. Windsurf (now Devin Desktop) reads mcp_config.json. For OpenClaw, use the CLI:

openclaw mcp add pagecrawl \
  --url https://mcp.pagecrawl.io/mcp \
  --transport streamable-http \
  --header "Authorization=Bearer YOUR_TOKEN_HERE"

Using Cursor, Copilot, Codex or Gemini CLI? Add the free PageCrawl skills so the agent picks the right tracking mode and asks before deleting:

npx skills add pagecrawl/skills -g

Full walkthrough

What your agent can do

Ask in plain language and the agent picks the right one of the 33 tools.

Set up monitors

Create monitors with the right tracking mode, frequency and alerts, apply templates, or discover pages across a whole site.

“Monitor example.com/pricing every hour and track the full page text.”

Read what changed

Changes since any date across every workspace, with AI summaries, importance and the exact diff.

“Summarise everything that changed across my monitors since last Monday.”

Check current values

Current prices, stock and other values for one monitor or up to 50 in a single call.

“What's the current price on my Amazon product monitor?”

Organise in bulk

Tag, move, edit and pause monitors, sort them into folders, and mark changes seen.

“Tag every monitor tracking amazon.com with 'competitor' and 'ecommerce'.”

Fix broken monitors

Find failing monitors, test a selector or tracking mode, look at the last screenshot, and run a one-off check.

“Are any of my monitors failing? Show me the ones with errors and what the issue is.”

Reports and webhooks

Schedule digest reports, connect webhooks, and pull workspace summaries and statistics.

“Set up a weekly digest of everything in the Competitors folder.”

What it can track
Full pageReaderPriceFeedSpecific textSpecific numberAI extractPDFJSON pathSEOHTMLJavaScriptVisualText presenceLinksWHOIS

Pages behind a login work too: the agent attaches a login saved in PageCrawl, and credentials never pass through the chat.

Secure and in your control

Sign in, no keys in the chat

Claude, ChatGPT, Cursor and VS Code sign in with OAuth on first connect. Clients that need an API token get one you can revoke any time from Settings, then API access.

Only your workspaces

The agent opens the same workspaces you see in PageCrawl and nothing else. Add someone to one workspace and their agent only reaches that one.

Asks before anything permanent

The server tells the agent to confirm with you before deleting monitors, clearing history or marking every change as seen.

Logins stay out of chat

The agent can attach a site login saved in PageCrawl to a monitor, but creating or editing a login is only possible in PageCrawl itself.

What teams using MCP say

“We query 60+ regulator pages from Claude without a single webhook. The receipts are signed and timestamped.”

Compliance officer, London-based asset manager

“Cursor + PageCrawl MCP is how my pricing agent reads the live web. It's the only monitor with a tool the LLM understands natively.”

Founder, AI agent platform

“Auth'd page monitoring from chat is the entire reason we picked PageCrawl over another well-known monitoring service.”

Procurement lead, UK NHS trust

Frequently asked questions

Connect PageCrawl to your agent

Sign up free, then add the connector or paste the config block.