shouldivibecodeit

Should I vibe codeHazel?

Watch selected folders and apply transparent file rules with previews and logs

A file rule with a bad glob doesn’t warn you. It quietly moves your tax returns somewhere.

?

Their verdict, the Hazel licence price and the build-time estimate come from their entry, MIT-licensed. Checked 2026-08-03.

Can you build it?asked by canivibecodeit.com ↗YESone-shottable · one sitting
?

Our verdict, the regret score and everything below it. Editorial and unsponsored — nobody can pay to be moved.

Should you ship it?asked by usDEMO ONLYvibe the v0, throw it away.

The honest answer

why the verdict is what it is

Watching a folder and applying rules is a fifty-line script. The problem is that the script’s mistakes are silent, immediate and applied to every matching file before you notice the pattern was wrong.

What actually breaks

not "if". the specific failures.

  • A glob that matches more than you intended, applied instantly to everything
  • A rule that fires on partially-written files while they are still downloading
  • Two rules that disagree, moving a file back and forth forever
  • Silent operation — you find out weeks later that something has been quietly filing your documents wrong
and then, at 3am

You are looking for last year’s tax return. It is not in Documents. It is not in the folder the rule was supposed to put it in either, because the rule matched on a substring that also appears in a client’s name, and eleven months of paperwork has been sorted into a directory you have never opened. Nothing errored. Nothing was logged. It just worked, confidently, on the wrong pattern.

Is that you?

the verdict is a default, not a law

ship it if
  • Rules only ever move files, never delete them
  • Every action is logged somewhere you will actually look
  • You have run it in dry-run mode over your real folders first
don’t ship it if
  • Any rule can delete, and there is no trash
  • It watches a folder that syncs to cloud storage, where a rename becomes a re-upload

If you build it anyway

the checklist, then the prompt that enforces it

  1. Move to trash, never unlink. The OS trash is a free undo and you should use it.
  2. Dry-run is the default. Live mode needs an explicit flag every time.
  3. Log every action with source and destination paths, to a file you keep.
  4. Debounce on file stability — do not act on a file whose size changed in the last few seconds.
  5. Cap actions per run and bail out loudly if a rule suddenly matches far more than usual.
  6. Never let two rules claim the same file. Detect the conflict and stop.
the guardrail prompt
I want a folder-watching file automation tool. The danger here
is silent wrongness, so:

1. Dry-run is the default mode and prints a table: file, matched rule,
   intended action, destination. Live mode requires an explicit --live flag on
   every invocation.
2. Never call unlink. Deletion means moving to the OS trash, always, with no
   configuration option to change that.
3. Before acting on a file, confirm it is stable: the size and mtime must be
   unchanged across two checks a few seconds apart. Do not act on partial
   downloads.
4. Append every action to a log file with an ISO timestamp, the rule name, the
   source path and the destination path. Then write an 'undo' command that
   reads that log backwards and reverses moves.
5. Add a sanity cap: if a single run would act on more than N files (default
   20), stop and print what it would have done instead. A rule that suddenly
   matches everything is a bug, not a busy day.
6. Detect overlapping rules at load time — if two rules could match the same
   file, refuse to start and tell me which ones.
7. Rules live in a plain config file I can read and diff. No binary state, no
   hidden database.
8. Add a --explain flag that takes a path and tells me which rules would match
   it and why, without doing anything.
paste this before you build — not after something breaks23 lines · 1287 chars

That one keeps you out of trouble. For the prompt that actually builds it, canivibecodeit.com has one.

their build prompt ↗

Or don’t build it

the boring option, and the way back out

just pay for it

Not really necessary here — this is a good build. The paid tools mostly win on the rule-editing interface, which matters less when you are the only person writing rules.

your exit plan, if you already built it

The action log is the exit plan. As long as every move is recorded with both paths, you can reverse the entire history of the tool with a script, which is the difference between a bad rule being an annoyance and being a loss.

prior art · someone already did this
Espanso

Active open-source cross-platform text expansion utility with system integration.

Questions

Why is a file organiser riskier than the tools rated below it?

Because it acts without being watched, its mistakes are silent, and by the time you notice, the pattern has been applied hundreds of times. Most tools fail loudly. This category fails quietly, which is worse.

did you build it?

Every week, someone ships something they shouldn’t have.

New verdicts, the worst thing that landed in the trap, and the occasional incident report. No other email, ever.

also on the regret index
Alfred PowerpackSHIP IT

A launcher is a fuzzy search over things you already have. Nothing here can hurt you.

BartenderSHIP IT

Hiding menu bar icons is not a subscription-shaped problem.

BetterTouchToolSHIP IT

Gesture bindings are a config file with opinions. Build yours.

last reviewed 2026-08-03 · verdict is editorial and unsponsored · shared entry data from canivibecodeit under MIT · not legal advice