// docs / connect-claude

Connect WebToMCP to Claude Desktop in 3 minutes

This tutorial walks you through giving Claude Desktop live, queryable access to any website using WebToMCP — index a public site, copy one URL, paste it into Claude's config. Total time: about three minutes. Free plan covers it; no credit card.

What you'll need

Claude Desktop installed — download it from Anthropic if you haven't.
A WebToMCP account (free) — sign up with Google.
A URL to index — your docs site, a wiki, an e-commerce store, anything public. Examples that work great: https://docs.lemonsqueezy.com/, https://tailwindcss.com/docs, https://hono.dev/docs.

Step 1 — Create a knowledge base

Sign in at app.webtomcp.net.
Click + New knowledge base.
Paste your URL (e.g. https://docs.lemonsqueezy.com/) and click Add.
Wait 1-10 minutes for crawling + indexing. The dashboard polls the progress live; you'll see pages count up.

Tip — path-prefix scoping: if you submit https://polar.sh/docs/introduction we only crawl URLs under /docs/. We won't bleed into /pricing or /blog on the same host. Keeps each knowledge base focused.

Step 2 — Copy your MCP endpoint URL

Once the knowledge base shows ready, click into it. The detail page shows your MCP URL — it looks like:

https://mcp.webtomcp.net/batches/<your-batch-id>/mcp?key=wmcp_<your-key>

Click Copy. Both the batch id and the URL key are scoped to your tenant — anyone with this URL can query the endpoint, so treat it like a personal access token.

Step 3 — Add it to Claude Desktop

The fastest path is the CLI:

npm i -g webtomcp-cli
wmcp connect claude

The CLI walks you through, writes the entry into Claude Desktop's config file, and tells you to restart Claude.

Doing it by hand

If you'd rather edit the config directly, open Claude Desktop's config.json:

macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
Windows: %APPDATA%\Claude\claude_desktop_config.json

Add (or merge) an entry under mcpServers:

{
  "mcpServers": {
    "webtomcp-docs": {
      "command": "npx",
      "args": [
        "-y",
        "mcp-remote",
        "https://mcp.webtomcp.net/batches/YOUR-BATCH-ID/mcp?key=wmcp_YOUR-KEY"
      ]
    }
  }
}

Replace the URL with the one you copied in Step 2, save, then quit and reopen Claude Desktop.

Step 4 — Verify it's working

In Claude Desktop, click the + or hammer icon — you should see webtomcp-docs listed as a connected MCP server. Now ask Claude a question that only the indexed site would know:

"What's the refund policy on this docs site?"

Claude should return an answer with source URLs from the indexed pages. If it does, you're done.

Troubleshooting

Claude doesn't show the server

Quit Claude Desktop fully (Cmd-Q / right-click tray → Quit, not just close window) and reopen. The config is read once at startup.

"Failed to connect to MCP server"

Double-check the URL has both the batch id and the ?key=... query string. The key is required for non-public knowledge bases. Re-copy from your dashboard if unsure.

Crawler is stuck at 0 pages

Some sites block crawlers in their robots.txt, return Cloudflare challenges, or 401 unauthenticated requests. The dashboard will surface the error inline once the timeout triggers. For JS-heavy sites, flip Enable JS rendering on the batch (metered: 5 pages/mo on Free, 100 on Hobby).

Answers feel stale

WebToMCP re-crawls weekly by default. Click Re-crawl in the dashboard to force an immediate refresh.

What's next

Connect more clients: wmcp connect chatgpt, wmcp connect cursor, wmcp connect gemini, wmcp connect codex.
Index multiple sites — Free plan covers 1 knowledge base, Hobby ($9/mo) covers 5.
Privacy & data custody — how we store the crawled content.

Stuck? Email developer@webtomcp.net — we reply fast in the beta.