Why they make no network requests
Both pages are aimed at people who will open devtools to check, so neither loads a
webfont, an analytics script or a font from a third party, and neither sends your input
anywhere. That is why they look plainer than the rest of this site. Once a page has
loaded you can turn the wifi off and it still works, and I recommend trying exactly that
before you paste anything real into either of them.
What these are not
Neither tool makes a judgment call for you. The appendix builder formats what you give it,
and the sanitizer does a first pass at removing identifying detail. The sanitizer in
particular is not a guarantee. It finds things shaped like a hostname or an
address; it cannot find a client named only in the middle of a sentence unless you tell it
the name. Read the output before you send it anywhere.
I am a developer, not an incident responder. These were built by reading published incident
reports and the shapes they print, not from case work. That is a real limit: they know the
shapes those reports use and nothing else. Where one of them meets your notes and gets it
wrong, hello@benchnotes.io reaches a person rather
than a queue.
What is next
There is a short study running on how incident reports actually get written, at
benchnotes.io/survey. It takes about six minutes, there is no pricing
question in it, and the results get posted publicly. If you want the next tool to be the one
you actually need, that is where to say so.