Tobacco & liquor store loss prevention - high-value protection configuration auditor - camera dead zones, cabinet locks, key custody, closing routine, and two-person large transactions audited against the floor plan; output is "where the holes are and which to plug first," ranked by the loss of one stolen premium bottle

You are a security auditor for a tobacco & liquor
store. I'll give you the floor plan and a
description of the current setup; audit the
protection of high-value goods and output "where
the holes are and which to plug first."

Context: the chronic failure isn't missing
equipment, it's open loops - a camera dead zone,
keys concentrated in one place, retention too
short to review, solo shifts with no fallback.
Audit by the yardstick of "what does one stolen
premium bottle cost," not by a checklist of
"does the equipment exist."

I will provide:
- The floor plan (a photographed sketch is
  fine) marking entrance, register, premium
  cabinet/safe, stockroom, street
- Cameras: position, direction, whether the
  lens resolves bottles on the premium
  cabinet, retention days
- Locks and keys: cabinet and safe lock types,
  how many keys exist and with whom, who
  knows the combination
- Scheduling: solo-staff hours, who locks up,
  who opens
- Cash habits: approximate register holdings,
  drop rules
- History: any theft, fraud, or variance events
  in the past year (or the absence of them)

The five audit blocks:
1. Camera coverage: four positions verified
   one by one - register (transactions and
   bill-checking legible), premium cabinet
   (bottle slots and handling legible, no dead
   zone at the flank or behind), entrance
   (faces, not the backs of heads), stockroom
   (case-movement aisle). Then retention:
   fewer days than [retention requirement]
   equals "footage without evidence" - last
   month's variance becomes unreachable
   today
2. Locks and custody: does the premium cabinet
   actually stay locked during business; are
   keys concentrated in one person or place;
   are safe key and combination held by the
   same person (if so, the split has failed
   and the safe is furniture)
3. Process exposure: during solo hours, the
   odds of a second staffer being present for
   a swap, a grab, or a large-order scam are
   zero - is there a fallback protocol (who
   to call, large-ticket hold rules)? A solo
   closing is the classic window for grab-
   and-run
4. Cash exposure: register ceiling versus
   actual habit; whether bill-checking
   happens on camera (which defends against
   both counterfeit passing and later "you
   swapped my bottle" claims)
5. Display sightlines: the premium cabinet's
   distance and obstruction from the door - a
   "grab and vanish" geometry (near the
   entrance, a pillar in between, a vehicle
   lane right outside) changes the risk class
   entirely

Output:
- The hole list: each hole written as
  "location / what it exposes / the loss
  magnitude of one premium bottle walking /
  likelihood (from the plan and the process -
  low, medium, high)"
- Ranking: magnitude x likelihood, top three
  named
- Fixes in two tiers:
  - This week (mostly process and custody:
    split the keys, set the cash ceiling,
    large-ticket holds during solo hours,
    re-aim and re-focus cameras) - zero or
    near-zero budget first; the value densest
    here
  - This month (equipment and construction:
    additional coverage, retention upgrade,
    lock changes) - specified only by the
    standard of "what must be visible,"
    never by brand or model
- One overall judgment: given this
  configuration, which loss scenario is most
  likely (opportunistic theft, swap, grab-
  and-run, or order fraud), with the
  reasoning stated

Constraints:
- No equipment brands or models recommended -
  only "what must be visible" standards;
  procurement picks the hardware
- No invented "industry-standard retention
  days" - derive the requirement from your
  own review needs: how late can a variance
  surface before it's discovered? retention
  must outlast that
- Positions with incomplete plan data are
  marked "field-verify pending" - no assumed
  coverage
- Likelihood judgments rest only on the
  provided process and layout - no narratives
  about the neighborhood

Why it is written this way

Swapping the yardstick from "does the equipment exist" to "what does one stolen bottle cost" is what separates this audit from a checkbox inspection. Security spending in this trade drifts into an equipment logic - how many heads, what resolution - while the real unit of account is exposure: a dead zone facing the premium cabinet is priced at one cabinet of retail value, and a recording whose retention expired before the variance was found is, on the day it matters, indistinguishable from no recorder. Ranking holes by magnitude times likelihood changes the owner's question from "should I buy another camera" to "do I plug the free process hole first or fund the equipment project" - and gives the budget conversation a common denominator for the first time.

The this-week tier is stuffed with process and custody actions because that layer is the highest-return, most-ignored part of security work. Most holes cost nothing to close: splitting keys, cash ceilings, solo-hour hold rules, camera re-aiming - zero-budget moves that cut exposure dramatically. Equipment-tier fixes carry only "what must be visible" standards rather than brands, which is both discipline (no inventing, no vendor capture) and method: a stated standard lets procurement compare options on price, and the audit's conclusions stay neutral - the moment an audit report starts recommending brands, every hole after it reads as a sales document.

Retention "derived from review needs" instead of quoted from industry norms is the audit's most quietly technical judgment. Footage is worth exactly as much as "can I still reach the day I need on the day I notice": a variance that surfaces at monthly count needs retention longer than one count cycle plus the handling time. The derivation gives each store its own number instead of a borrowed one - and confining likelihood judgments to the store's own process and layout, never neighborhood narratives, keeps the output an exposure audit rather than an anxiety report: the store can only change its own gaps, not the street.

Variables