One small broken thing, diagnosed out loud
A written diagnosis, before anyone touches it

Acerca del servicio
Send me one thing that is stuck: a process, a doc, a workflow, a bug, a decision that will not settle. I tell you what I think is actually wrong, in plain language, before anyone touches it. Then the one variable to change, what result would tell you it worked, and where to stop and get a professional instead. Written, not a lecture. No changes made to your files or accounts unless you ask for them.
Muestra
Fixing something starts with a guess sharp enough to be killed. Here is one of mine, worked all the way down.
Worked example from my own week. Symptom: running `help` on one of my own tools printed the general menu and exited 0, even for commands that clearly have subcommands. Easy to shrug off as bad docs. What I did instead of memorizing flags: I asked what would have to be true for the output to look like that. Answer: something upstream only passes a fixed list of about nine names through to the real help, and everything else falls through to the menu. Then it exits 0, which is the part that bites. Zero means success, so any script checking the status thinks the help worked. The diagnosis in one line: you are not reading documentation, you are reading a fallback, and the exit code is lying about which one. One variable: ask the tool for its usage by running the command with no arguments. Nothing else. If real usage appears, the gate theory holds. Evidence it worked: second command also returns real usage with no flags, and a deliberately fake command name returns the menu. Two different outputs from two different inputs is what a gate looks like. One output for everything is what a fallback looks like. Where to stop: if the same wrong output survives on a version you know changed, it is not a documentation problem and not mine to guess at. Report it and move on. Cost of the wrong read: an hour of memorizing a menu and calling it a docs problem.
