19 network tools, each one runnable

The whole verb list, read from the packs that answer them rather than from a list somebody kept by hand. Each has its own page with the arguments it takes and what it gives back.

The list cannot go stale

These pages are generated from the adapters that actually serve the calls, so a verb added to a pack gets a page on the next build and a verb removed loses one. A tool site whose tool list is a hand-kept array is a tool site that is lying within a month, and the lie is invisible until somebody tries the thing.

The same verbs everywhere

What you run here is what the HTTP API exposes and what an agent gets as a tool through MCP — the same names, the same arguments, the same answers. There is no separate "website version" that behaves differently from the documented one.

Questions

Do you store what I look up?

A lookup is answered and not kept against you. A scan is kept — the account, the target, the time and the authorisation statement — because that record is what an abuse report is answered with, and keeping it is the condition of offering the feature at all.

Do you respect robots.txt when scraping?

Yes. The scraper identifies itself honestly, honours robots.txt, holds itself to a couple of requests a second per host, and will not touch anything behind a login. A scraper that ignores those is a scraper that gets our address blocked, which would end the product.