I've been working on an open source MCP (Model Context Protocol) server for
webcrawl data. Basically, the server offers advanced search and retrieval of
web pages/assets through a MCP-capable LLM. You point it at the web archives,
and can access search from within the LLM app.

I just recently added HTTrack support. If anyone is interested in trying this
out, the code is on github and I've put together a short tutorial on setting
it up.

Repo: 
<https://github.com/pragmar/mcp-server-webcrawl>

Setup: 
<https://youtu.be/HAVfvmrZjRk>