Free tools for the questions that come up when something is slow, unreachable or misconfigured. They report what was measured, explain what it means, and say what to check next — and they say when a measurement could not be made, rather than filling the gap with a plausible number. No account, and no sign-up to see results.
Throughput, latency, loss and the behaviour of a connection under load.
Whether a network will carry calls well, and what to change if it will not.
Records, resolution, signing, and who owns what.
The DNS and transport configuration that decides whether mail arrives.
Certificates, headers, exposure and reachability.
Response times and reachability from more than one place.
Arithmetic that is tedious by hand and instant in a browser.
8 tools are available today; 22 more are in build. Tools without a link are not finished — we would rather show a short list than send you to a page that does not work.
“Jitter: 45 ms” is a fact nobody can act on. What matters is that jitter at that level makes voice calls choppy, and that the usual causes are Wi-Fi interference, a congested uplink, or no traffic prioritisation. Every result here comes with what it means and what to check.
A tool that cannot measure something says so. Where a result is a hypothesis rather than an observation — which NAT behaviour is likely, which of several causes fits — it is labelled as one. Confident numbers invented to fill a gap are worse than an honest blank.
Each tool states what it tested, from where, and what it cannot see. A speed test measuring against a node in the next city and one across an ocean are different measurements, and you should be able to tell which you got.
Several of these run entirely in your browser and send nothing anywhere — the subnet calculator among them. Where a tool does need a server, the page says what it sends and why.