WebMCPIndexBot

If this crawler reached your site and you want it to stop, the fastest route is robots.txt. We honour it. If you would rather email us, write to contact@webmcpshield.com and we will remove your domain by hand.

What it does

WebMCPIndexBot measures how the web is adopting WebMCP — a browser API that lets a page register tools an AI agent can call. We publish the aggregate result at webmcpshield.com.

On a normal visit the bot performs one HTTP GET of your home page, reads the HTML, and leaves. It looks for the JavaScript that registers WebMCP tools. If a site declares none — which is almost all of them — nothing further happens.

What it does not do

How often

Most domains are visited once every 15 days. Sites that declare WebMCP tools are re-checked more often, because changes to those declarations are the thing we track. Requests to a single host are spaced at least one second apart.

How to block it

Add this to your robots.txt. It takes effect on the next visit.

User-agent: WebMCPIndexBot
Disallow: /

We fetch and honour robots.txt before requesting anything else, and we treat a disallow as final — the domain is skipped, not retried under a different name.

Blocking by User-Agent or by IP at your edge also works, and we will not attempt to circumvent it.

Identifying it

User-Agent: WebMCPIndexBot/0.1 (+https://webmcpshield.com/bot)

Nearly all requests come from a single address:

139.162.75.45

A small number of follow-up requests — the ones that load a page in a real browser to read its tool declarations — originate from other hosts we operate. If you see traffic identifying itself as WebMCPIndexBot from a different address and want it confirmed, email us and we will tell you whether it is ours.

Contact

contact@webmcpshield.com

Removal requests are actioned without argument. Please include the domain. We aim to reply within one business day.