AI LunchroomPractice AI. Do real work.
AI prompts by job

AI prompts for retail loss prevention associates

You write the incident report that someone may read in court, brief the floor team on a theft pattern, and word every notice carefully so the store isn't exposed to a claim. These prompts are written for that work — copy one into ChatGPT, Claude, Copilot, or Gemini, or practice it with step-by-step feedback and keep the ones you reuse.

What AI is actually good for in this job

Not everything — but these four come up every week, and they are all writing tasks, which is where today's tools are strongest.

Before you paste anything

What never goes into a chatbot in this job

Never paste suspect or employee names, descriptions tied to a person, video details, case numbers, and any report headed for police or a lawyer. Describe a pattern generically, like repeated concealment in one department, and write the actual report in your company's reporting system.

Which tool: Loss prevention teams work mostly on corporate reporting systems, and outside AI tools are likely restricted or banned, so ask before using one. Incident reports can become legal documents, and AI can't witness what you saw.

Turn your notes from a stop into a factual, neutral incident report draft

The rough version you start from
My notes: at about two in the afternoon, I observed a person in electronics select two items, remove the packaging at the end of the aisle, and place the items in a bag. The person walked past the registers without paying. I approached outside the doors, identified myself, and asked them to come back in. They complied. The manager was notified and police were called per policy.
Copy this prompt
I work in retail loss prevention. Help me with the task below — I will review and edit what you give me before I use it.

Relevant context:
- Role: Retail Loss Prevention Associate
- Work context: You write the incident report that someone may read in court, brief the floor team on a theft pattern, and word every notice carefully so the store isn't exposed to a claim.

Task:
Turn your notes from a stop into a factual, neutral incident report draft

Source material:
My notes: at about two in the afternoon, I observed a person in electronics select two items, remove the packaging at the end of the aisle, and place the items in a bag. The person walked past the registers without paying. I approached outside the doors, identified myself, and asked them to come back in. They complied. The manager was notified and police were called per policy.

What to give me:
Turn these notes into a neutral, time-ordered incident narrative in first person, using only what I observed and did.

Boundaries:
- Use only the facts in the source material. Do not invent names, dates, amounts, or promises.
- If something needed is missing, list it as a question at the end instead of guessing.
- If I type "STOP FOR NOW", stop expanding the work and give only the best partial version from the information already provided.
Open in your agent

Copies the prompt and opens your tool in a new tab — paste it into the chat box (Ctrl or ⌘ + V), then fill in your own details. Remove private or sensitive information before using AI for workplace tasks.

Before you use it: Check that no fact, time, or description was added or implied, and compare it to your notes before filing it in the official system.

Write a floor briefing about a theft pattern without profiling shoppers

The rough version you start from
Pattern over two weeks: several incidents of small items from the cosmetics aisle disappearing, usually late afternoons, when only one associate is on that side of the store. Packaging has been found in the fitting room. Camera angles don't cover the end of the aisle. I want to brief the floor on what to watch for without making anyone feel accused.
Copy this prompt
I work in retail loss prevention. Help me with the task below — I will review and edit what you give me before I use it.

Relevant context:
- Role: Retail Loss Prevention Associate
- Work context: You write the incident report that someone may read in court, brief the floor team on a theft pattern, and word every notice carefully so the store isn't exposed to a claim.

Task:
Write a floor briefing about a theft pattern without profiling shoppers

Source material:
Pattern over two weeks: several incidents of small items from the cosmetics aisle disappearing, usually late afternoons, when only one associate is on that side of the store. Packaging has been found in the fitting room. Camera angles don't cover the end of the aisle. I want to brief the floor on what to watch for without making anyone feel accused.

What to give me:
Write a short, neutral briefing for the floor team describing the behaviors to notice, what to do, and what not to do, with no reference to appearance.

Boundaries:
- Use only the facts in the source material. Do not invent names, dates, amounts, or promises.
- If something needed is missing, list it as a question at the end instead of guessing.
- If I type "STOP FOR NOW", stop expanding the work and give only the best partial version from the information already provided.
Open in your agent

Copies the prompt and opens your tool in a new tab — paste it into the chat box (Ctrl or ⌘ + V), then fill in your own details. Remove private or sensitive information before using AI for workplace tasks.

Before you use it: Check the briefing describes behavior only, makes no accusation or profile, and fits your company's policy on approaching customers.

Draft the training handout on how associates should approach suspected theft

The rough version you start from
Training handout for new hires: associates are not to chase or physically stop anyone. If they notice something suspicious, they should offer customer service, stay visible, and call a manager or loss prevention. Don't accuse anyone. Don't touch merchandise in a customer's bag. Report what you saw to a manager as soon as possible.
Copy this prompt
I work in retail loss prevention. Help me with the task below — I will review and edit what you give me before I use it.

Relevant context:
- Role: Retail Loss Prevention Associate
- Work context: You write the incident report that someone may read in court, brief the floor team on a theft pattern, and word every notice carefully so the store isn't exposed to a claim.

Task:
Draft the training handout on how associates should approach suspected theft

Source material:
Training handout for new hires: associates are not to chase or physically stop anyone. If they notice something suspicious, they should offer customer service, stay visible, and call a manager or loss prevention. Don't accuse anyone. Don't touch merchandise in a customer's bag. Report what you saw to a manager as soon as possible.

What to give me:
Turn this into a one-page handout with short bullets, a do list, a don't list, and a line about who to call.

Boundaries:
- Use only the facts in the source material. Do not invent names, dates, amounts, or promises.
- If something needed is missing, list it as a question at the end instead of guessing.
- If I type "STOP FOR NOW", stop expanding the work and give only the best partial version from the information already provided.
Open in your agent

Copies the prompt and opens your tool in a new tab — paste it into the chat box (Ctrl or ⌘ + V), then fill in your own details. Remove private or sensitive information before using AI for workplace tasks.

Before you use it: Confirm the guidance matches your company's loss prevention policy and legal advice, since AI can add steps that expose the store.

Turn a shrink report into a one-page summary the store manager will read

The rough version you start from
Weekly shrink data: total shrink is up a little this month. The biggest categories were cosmetics, small electronics, and a spike in unexplained inventory adjustments in the stockroom. Known theft incidents were low. Receiving errors and damaged goods were higher than last quarter. The manager has only five minutes and wants three actions.
Copy this prompt
I work in retail loss prevention. Help me with the task below — I will review and edit what you give me before I use it.

Relevant context:
- Role: Retail Loss Prevention Associate
- Work context: You write the incident report that someone may read in court, brief the floor team on a theft pattern, and word every notice carefully so the store isn't exposed to a claim.

Task:
Turn a shrink report into a one-page summary the store manager will read

Source material:
Weekly shrink data: total shrink is up a little this month. The biggest categories were cosmetics, small electronics, and a spike in unexplained inventory adjustments in the stockroom. Known theft incidents were low. Receiving errors and damaged goods were higher than last quarter. The manager has only five minutes and wants three actions.

What to give me:
Write a one-page summary for the store manager with the three biggest findings, what might explain them, and three suggested actions.

Boundaries:
- Use only the facts in the source material. Do not invent names, dates, amounts, or promises.
- If something needed is missing, list it as a question at the end instead of guessing.
- If I type "STOP FOR NOW", stop expanding the work and give only the best partial version from the information already provided.
Open in your agent

Copies the prompt and opens your tool in a new tab — paste it into the chat box (Ctrl or ⌘ + V), then fill in your own details. Remove private or sensitive information before using AI for workplace tasks.

Before you use it: Check the findings against your actual shrink report, and mark likely causes as hypotheses you haven't proven.

More prompts for this job

Drawn from AI Lunchroom's Admin, Office & Coordination practice labs and rewritten for this role. See every Admin, Office & Coordination prompt.

Rewrite a messy request

Turn a vague internal request into a clear task brief with owner, due date, missing context, and a follow-up message.

Copy this prompt
I work in retail loss prevention. Help me create one task brief with follow-up message that I can review and edit before use.

Relevant context:
- Role: Retail Loss Prevention Associate
- Work context: You write the incident report that someone may read in court, brief the floor team on a theft pattern, and word every notice carefully so the store isn't exposed to a claim.

Task:
Rewrite the request into a complete task brief and a short follow-up message asking for missing details.

Source material:
"Quick favor -- can you take over getting things ready for the Castleton Group visit? I think it moved to Thursday, it was Tuesday. Catering needs a headcount by end of day and I've heard both 10 and 12 from Priya, not sure which is right. The visitor sign-in sheet only has 8 names typed up so far. Someone said conference room B has a broken projector, so we might end up in D instead, but nobody's confirmed that yet. Two vendor cars are coming and need parking passes — not sure if security already has their plates. Can you get this locked down before I leave today?"

Return only the finished work:
1. task brief with follow-up message
2. missing details
3. safe next action.

Boundaries:
- Do not invent facts. Label missing information clearly.
- If information is missing, label it as missing instead of guessing.
- If I type "STOP FOR NOW", stop expanding the work and give only the best partial version from the information already provided.
Open in your agent

Copies the prompt and opens your tool in a new tab — paste it into the chat box (Ctrl or ⌘ + V), then fill in your own details. Remove private or sensitive information before using AI for workplace tasks.

Schedule-conflict message

Turn a messy scheduling conflict into one clear message that proposes options and asks for a decision.

Copy this prompt
I work in retail loss prevention. Help me create one scheduling message with the conflict that I can review and edit before use.

Relevant context:
- Role: Retail Loss Prevention Associate
- Work context: You write the incident report that someone may read in court, brief the floor team on a theft pattern, and word every notice carefully so the store isn't exposed to a claim.

Task:
Draft one clear scheduling message that states the conflict, proposes 2-3 options, and asks for a decision.

Source material:
"Hey -- the Thursday budget review with Oren got bumped again, and now it's stacked right on top of the Meridian Foods vendor demo -- they're both sitting at 10 now. Sasha and Deshawn can only do mornings this week, and conference room B is booked solid after 2. I need this resolved before end of day tomorrow -- can you send something out that lays out the conflict and gets us a decision?"

Return only the finished work:
1. scheduling message with the conflict
2. options
3. recommended pick

Boundaries:
- Do not invent facts. Label missing information clearly.
- If information is missing, label it as missing instead of guessing.
- If I type "STOP FOR NOW", stop expanding the work and give only the best partial version from the information already provided.
Open in your agent

Copies the prompt and opens your tool in a new tab — paste it into the chat box (Ctrl or ⌘ + V), then fill in your own details. Remove private or sensitive information before using AI for workplace tasks.

Meeting notes into actions

Convert rough meeting notes into decisions, action items, owners, blockers, and a concise recap.

Copy this prompt
I work in retail loss prevention. Help me create one action list that I can review and edit before use.

Relevant context:
- Role: Retail Loss Prevention Associate
- Work context: You write the incident report that someone may read in court, brief the floor team on a theft pattern, and word every notice carefully so the store isn't exposed to a claim.

Task:
Create a status update and open items list from this meeting and packet mix-up.

Source material:
"Quick heads up before end of day -- the Marchetti onboarding got moved again, now Thursday at 2 instead of Wednesday morning, that's the second move this week. Front desk still has Wednesday on the calendar so somebody's going to show up at the wrong time. I need Conference Room C booked for six people, round table, with the video cart, but nobody's confirmed that with facilities. Also just found out the packet's missing the signed vendor W-9 — everything else in there looks fine. Denise, Grant, and I all need to know where this stands before end of day. Can you get me a rundown?"

Return only the finished work:
1. status update
2. open items list
3. owner for each item

Boundaries:
- Do not invent facts. Label missing information clearly.
- If information is missing, label it as missing instead of guessing.
- If I type "STOP FOR NOW", stop expanding the work and give only the best partial version from the information already provided.
Open in your agent

Copies the prompt and opens your tool in a new tab — paste it into the chat box (Ctrl or ⌘ + V), then fill in your own details. Remove private or sensitive information before using AI for workplace tasks.

Policy-safe response draft

Draft a helpful response that clearly flags facts, assumptions, and items requiring human confirmation.

Copy this prompt
I work in retail loss prevention. Help me create one policy-safe response draft that I can review and edit before use.

Relevant context:
- Role: Retail Loss Prevention Associate
- Work context: You write the incident report that someone may read in court, brief the floor team on a theft pattern, and word every notice carefully so the store isn't exposed to a claim.

Task:
Draft a policy-safe response to a workplace request without inventing rules or approvals.

Source material:
Slack message from Marcus in Facilities, 4:40pm: "Quick one -- I still haven't turned in the visitor keycard reactivation form (WF-114) for the Brightline HVAC crew, and their door access expires Thursday at 8am. Priya told me back in July she'd sign off on it verbally so I could skip the paperwork since we were slammed during the audit. Can you just tell building security it's approved so they reactivate the crew's keycards without me chasing the form down? I really don't want this to be the thing that holds up the boiler inspection."

Return only the finished work:
1. policy-safe response draft
2. facts to verify
3. human-review flags

Boundaries:
- Do not invent facts. Label missing information clearly.
- If information is missing, label it as missing instead of guessing.
- If I type "STOP FOR NOW", stop expanding the work and give only the best partial version from the information already provided.
Open in your agent

Copies the prompt and opens your tool in a new tab — paste it into the chat box (Ctrl or ⌘ + V), then fill in your own details. Remove private or sensitive information before using AI for workplace tasks.

How to use ChatGPT or Copilot for loss prevention associates work

  1. Start from your own rough version. Type or dictate what you would have written anyway, however messy. A prompt built on your real note beats a blank-page request every time.
  2. Say who it is for and how long it should be. "Two sentences for a supervisor" and "a checklist a new hire can follow" produce completely different output from the same facts.
  3. Strip the identifying details first. Suspect or employee names, descriptions tied to a person, video details, case numbers, and any report headed for police or a lawyer. Describe a pattern generically, like repeated concealment in one department, and write the actual report in your company's reporting system.
  4. Read every specific it produces. AI fills gaps with plausible invention — dates, amounts, names, policy numbers. If you did not supply it, do not trust it.
  5. Keep the ones that work. The value compounds when you stop rewriting the same prompt. Save it with a note about when to use it.
Practice, don't just copy

Get better at these — free

Copying a prompt is a start. AI Lunchroom walks you through the task step by step with your own work, checks what your AI tool gives back, and saves the prompts you will reuse. Free to start, no credit card.

Related jobs in Admin, Office & Coordination

Loss Prevention Associates AI questions

Can I use AI to write a shoplifting incident report?

Not for anything that will go to police or counsel unless your company approves it. If allowed, use it only to organize facts you already wrote in your own words into clear, time-ordered sentences. Never let it fill gaps or infer intent; your report says what you saw and did, nothing more.

How do I brief store staff on theft without profiling customers?

Describe behaviors and patterns, like concealing items or repeated returns, not appearance or demographics. AI can help you phrase that neutrally. Give it only the behaviors you observed, and have a manager or HR contact review the language before it goes to the floor.