Convenience store backroom manager - backroom location labels and find-it index tool: location codes fixed as zone-aisle-level, labels in type sizes readable at one meter in a dim backroom, a door-side quick index in both category and first-letter order, managing "where it is" and never "how many" so it never fights the POS inventory

You are a coding agent. Build a single-file HTML
"backroom location labels and find-it index tool"
for a convenience store backroom manager: move
"where things are" out of people's heads and into
printable labels and an index sheet. Runs locally,
no network, no external dependencies.

Positioning (design to this; do not build
inventory software): a convenience store backroom
has no warehouse system; location knowledge lives
in veteran employees' memory. This tool solves
exactly one problem - anyone (night-shift newbie,
cover staff, count temps) can find where an item
sits within 30 seconds. It deliberately does not
do quantities: no counts, no stock levels, no
ordering - those belong to the register system.
Two ledgers always fight; with one ledger, the
system is the only source of truth.

Two input blocks:
1. Location definitions: zones (letters, e.g.
   A dry / B direct-delivery / C fresh staging /
   D supplies), aisles per zone, levels per
   aisle. The code scheme is fixed:
   zone-aisle-level, so A-03-2 = zone A, aisle
   3, level 2
2. Items and locations: item name (aliases
   supported, e.g. "lighters / fire starters"),
   default location, optional "commonly
   misplaced at" (history: where this item keeps
   getting dropped by mistake)

Three outputs:
1. Label print pages:
- One large label per lane entrance or rack
  post: zone letter in display type (readable
  at one meter after printing - at least a
  quarter of an A4 sheet), the location range
  in that zone, and the categories stored there
- One small label per aisle side: aisle number
  plus up to three lines of what lives on each
  level
2. Find-it quick sheet (print, inside the
stockroom door):
- Two indexes side by side: left column by
  category (fresh / tobacco / cleaning /
  supplies - categories editable), item to
  location; right column alphabetically by
  first letter, item to location - the category
  index serves veterans, the letter index
  serves new hires
- A separate "commonly misplaced" section:
  items with a record only, each line reading
  item - check the default location first,
  then the misplacement spot
3. Maintenance view:
- Add and edit items, locations, misplacement
  spots; every change logged in "last ten
  changes" (time, what changed) so "who moved
  it and when" is answerable
- Data in browser localStorage with JSON
  export / import - survives a machine change
  or reinstall
- A demo store preloaded (two zones, ten items,
  two misplacement records) so all three
  outputs are visible the moment it opens

Hard constraints:
- The zone-aisle-level scheme is not
  configurable - a configurable scheme means
  every store speaks its own dialect that cover
  staff and temps can never learn; a fixed
  scheme keeps labels usable even away from the
  tool
- Printed labels must stand alone: no reliance
  on color (black-and-white printing loses
  nothing), no QR-code jumps (nobody in a
  backroom has a scan setup)
- The quick sheet fits its page; on overflow it
  paginates with page-order navigation

You must not:
- Build quantity tracking, order suggestions,
  or date alerts - "how many" belongs to the
  register system alone; the moment this tool
  touches quantities it becomes a second ledger
- Build scanning, Bluetooth, or any network
  feature
- Add accounts or external libraries

Why it is written this way

"Where, never how many" is the load-bearing boundary of the whole prompt. Backroom tools most often die by accreting quantity features: log a case today, forget to decrement tomorrow, and by day three the tool and the register system disagree - and whenever two ledgers exist, people trust the one at hand. Banning quantities outright confines the tool to what the POS genuinely cannot do (the POS knows how many; it does not know that shelf level three holds lighters rather than batteries), and two tools that each own one question coexist.

The zone-aisle-level scheme is fixed and non-configurable as a deliberate trade of flexibility for usability. Let stores customize codes and each invents its own dialect that cover staff and count temps must relearn at every door; with one fixed scheme, labels stay valid away from the tool - a temp who has never seen it still knows what to do in front of a tag reading A-03-2. The dual category-and-letter index on the quick sheet follows the same logic: veterans think in categories ("where's cleaning"), new hires only remember item names ("where's the bleach") - one sheet serves both.

"Commonly misplaced" looks like a minor feature but is the structuring of tribal knowledge. Every store has a few spots where things just get dropped by habit; that knowledge never enters a system, and new hires only discover it through repeated failed hunts. Letting the backroom manager record misplacements turns the quick sheet into an executable search route - default location first, known misplacement spot second - instead of "ask somebody."

Labels that stand alone (black-and-white safe, no QR codes) and exportable data are two faces of one requirement: the print artifacts must be fully self-sufficient once they leave the digital world - a dim, dirty backroom with no scanner has to absorb the label's information in one glance; meanwhile the digital data (localStorage) must be portable, because a slotting sheet that dies with a cleared browser cache will never be maintained a second time. The preloaded demo store continues the same principle: the tool's value shows within ten seconds of opening, and an empty sheet demonstrates nothing.

Variables