Complete AI Training
Sign inGet my AI kit

Your job's AI kit

Get your AI kit

Tell us who you are and what you do. We show you your kit right away and email you the link: skills, prompts, AI agents, MCP servers and courses for your job.

500+ jobs ready, and we make a kit for any other job. No payment needed to look.

Share

AI agent for pharmacists

Controlled Substance Count Reconciliation Agent

Counts match records, and every difference is explained

Controlled Substance Count Reconciliation Agent: what goes in, what the agent does and what you get

What it does

Controlled drugs must be counted and reconciled against purchase, dispensing and waste records, and unexplained differences carry legal weight. This agent reads the count, the dispensing or administration log, purchase invoices and waste records, and calculates expected on-hand for each drug. When the count does not match, it lists transactions since the last clean count and looks for likely causes, like unrecorded waste or an unposted invoice. Staff recount and add missing records, and the agent checks again. Any difference that no record explains is marked unexplained and a recount is requested. Unresolved differences go to the responsible person for reporting. The pharmacist signs off every reconciliation and any loss report. Edge case: a partial vial with witnessed waste that was never logged is flagged, not assumed.

How it works

Follow the arrows from top to bottom. The orange dashed arrow is the loop: when a check fails, the agent goes back and tries again.

Start and resultWhat it doesA check on its own workWaits for your OKGoes back and retries
Yes, continueYes, continueApprovedNoNo 1 STARTS WHEN Count entered 2 USES A TOOL Read dispensing or administration logs 3 USES A TOOL Read purchase invoices and waste records 4 DOES Calculate expected on-hand for each drug 5 CHECKS THE RESULT Does the count match expected on-hand? If not: list transactions since the last clean count andlook for missing waste or unposted invoices. Back tostep 4. 6 DOES Draft discrepancy note with likely cause 7 CHECKS THE RESULT Is every difference explained by a record? If not: mark it unexplained and ask for a recount. Backto step 2. 8 YOU APPROVE Responsible person signs off 9 RESULT Reconciliation record
Read the steps as a list
  1. Count entered
  2. Read dispensing or administration logs
  3. Read purchase invoices and waste records
  4. Calculate expected on-hand for each drug
  5. Does the count match expected on-hand?If not: list transactions since the last clean count and look for missing waste or unposted invoices. Back to step 4.
  6. Draft discrepancy note with likely cause
  7. Is every difference explained by a record?If not: mark it unexplained and ask for a recount. Back to step 2.
  8. Responsible person signs offThe agent waits here for your OK.
  9. Reconciliation record

How it decides

Expected minus counted equals difference. Any difference is investigated.

  • Any difference triggers review
  • Waste needs a witness
  • Unresolved after 24 hours escalates

Make it yours

Every agent is a starting point. You choose these settings for your own situation.

  • Count frequency
  • Drugs in scope
  • Escalation time
  • Report format

What keeps you in control

It always asks you first

  • Sign-off
  • Loss reports

Hard limits

  • No log edits by the agent
  • Loss reporting by a person

It stops when

  • Done: matched
  • Stop: possible loss, report per rules

Set it up

We guide you through the set-up, step by step

Members get the full set-up guide for this agent. No technical skills needed: you copy, paste and upload.

10 minto set it up in your AI
5 AIsChatGPT, Claude, Copilot, Gemini, Grok
  • One set of instructions to paste into your AI, with the clicks for ChatGPT, Claude, Microsoft 365 Copilot, Gemini and Grok
  • The agent then walks you through connecting your own data, one source at a time
  • A downloadable copy with the flow chart, the rules and the full guide
Get access to this agent

An example run

What happensOn the Friday count, oxycodone 5 mg showed 412 tablets against an expected 442. The agent listed transactions since the last clean count on Monday. It found an invoice for 30 tablets received Wednesday but returned the same day as damaged, with no return record posted. After the return form was found and entered, the count matched. The pharmacist in charge signed off on the reconciliation.

More agents for pharmacists