Hire an AI clinic referral handover reviewer

Everything it can hand back

Your task sets the actual outputs. The specialist picks the methods that cover them.

• A referral status handover, with each referral shown at the furthest step your records prove: sent, received, booked or completed
• A list of missing confirmations, with the last date anything was confirmed
• Draft follow-ups for you to send, written to your clinic's rules
• A short list of items that need a clinician or the clinic contact
• MISSING lines for each gap, and a record to start the next review from

How it does the job well

• Turn every request into owned actions: who, what and by when.
• Work from the real calendar, notes and lists. Never invent meetings, people or dates.
• Put the most important decision or deadline first, then the details.
• Prepare drafts and proposals. Sending, booking and changing records waits for approval.

Mistakes it avoids

• Repeating old recommendations as new findings.
• Treating a status word, a confident note or an empty export as proof.
• Keeping personal records in memory or logs.
• Describing installation as a running automation.
• Losing commitments buried in long notes.
• Mixing confirmed facts with guesses about other people's plans.

What a good result looks like

An example, not your task: A referral coordinator is leaving or going on leave, and a colleague takes over the list.

• Referral handover. Every referral at the furthest step your records prove, with the proof and the next confirmation needed.
• Missing confirmations. The referrals with no proof of receipt, booking or completion, and the last date anything was confirmed.
• Follow-up drafts. A draft message for each missing confirmation, written to your clinic's rules. None is sent.
• Clinician contact list. Items that need the clinic contact, with their role. The agent adds no clinical judgment.
• Accepted follow-ups. Follow-ups that someone has accepted, with the owner and the date.
• MISSING lines and CHECKS MISSED. Every fact the records did not give and every check that could not run.

How the work is checked

• No referral is shown as received, booked or completed without its own proof.
• No file contains a name, date of birth, address or clinical detail.
• No urgency, priority or outcome appears anywhere.
• Every draft follow-up uses the user's own rules, or says MISSING.
• States which checks passed, failed or could not run, and names missing inputs.

These checks have not been run against a model yet. Your AI uses them to check its own work.

Capabilities

• Read the last review. Loads the saved record: referral reference, administrative status, confirmation date and accepted follow-up. A first run builds a baseline with you.
• Keep only what is needed. The Source note summariser pulls out the reference, the status and the confirmation evidence. It leaves out names and clinical details, and flags any it finds in your records.
• Find the proved step. Sent needs proof of dispatch. Received needs a receipt. Booked needs a booking record. Completed needs a completion record. Silence means not confirmed.
• Write the handover. The Prepare a referral administration handover workflow runs the Source note summariser, then the Handover note writer. Pending confirmations and clinician contact points come first.
• Draft and record follow-ups. Writes one draft follow-up for each missing confirmation, to your clinic's rules. When someone accepts a follow-up, the Executive action register keeper records the owner and the date.
• Check and save. Checks that no urgency, outcome or confirmation is invented. Saves the files, a run log and the new record, and keeps the old record.

How the role works

1. Checks your task, context and available tools.
2. Picks the methods needed and prepares a work plan.
3. Carries out supported tasks, with each method's limits and review gates.
4. Checks the real deliverables and corrects supported faults.
5. Hands back files, evidence and unresolved gaps.

Your specialist's working setup

Download the base specialist without a task form or account. README.md and SETUP.md explain how to use the folder. SOUL.md defines its scope. INTAKE.md asks missing questions inside your AI. PLAYBOOK.md, SKILLS.md, BEST_PRACTICES.md, CHECKS.md and HANDOFF.md guide the work and its checked deliverables.

Complete associated skills, supporting files and licences are included. Choose only the methods needed for the actual task. Keep context, working files and accepted feedback in your AI project.

Use this specialist through MCP or the API

Call prepare_specialist with role_id referral-administration-evidence-review-assistant only. No task is required to load it. Your AI asks what job you need done, what you want back and which relevant context it can use.

For task-led hiring, call find_specialist, explain the suggested specialists and let the user accept or change the team.

Complete ZIP: Download base specialist. API: GET /api/v1/specialists/referral-administration-evidence-review-assistant/pack. The host AI supplies its own model, tools and execution; a pack does not connect accounts or run a task.

Inputs and helpful context

The specialist starts from the task brief. Only active methods require their inputs.

For the primary method, when used

• Referral status records with a reference number, a status and dates (from your referral log or practice system; remove names and clinical details first)
• Proof of sending and receipt, such as a dispatch confirmation, a receipt notice or a portal status
• Booking confirmations, where there are any (optional)
• Your clinic's approved follow-up rules, such as how long to wait before chasing and who to contact (from your procedures)
• The name or role of the incoming administrator, and the contact for clinical questions
• Last review's record (optional, from the second run on)

Helpful context (optional)

• Your referral log export, with identifiers removed.
• Receipt notices or portal status pages.
• Your follow-up and escalation procedure.

Tools and access

• A capable chat AI with file support: Keeps the record between reviews. Only add records your clinic allows you to share with an AI tool.
• A chat AI that reads files: It reads a status export and your rules together. Check your clinic's rules before you share records with any AI.

Functions and deliverables

• Clinic referral handover reviewer: Track clinic referral records. Get a handover with missing confirmations, clear status and follow-up drafts.
  • A referral status handover, with each referral shown at the furthest step your records prove: sent, received, booked or completed
  • A list of missing confirmations, with the last date anything was confirmed
  • Draft follow-ups for you to send, written to your clinic's rules
  • A short list of items that need a clinician or the clinic contact
  • MISSING lines for each gap, and a record to start the next review from
• Clinic referral administration handover planner: Share approved referral records and clinic rules. Get handover notes, open status checks and follow-up drafts for the next admin.
  • Pending confirmations, listed first
  • Each referral's state (sent, received, booked or completed), with proof and date
  • Any line that needs a clinician, copied as written
  • A draft follow-up for each unconfirmed referral, marked not sent
  • Contacts, and a list of gaps
• Action item tracker: Share meeting notes or email threads. Get action items with owners, deadlines and unresolved commitments.
  • action-register.csv: id, task, owner, due_at, timezone, status, depends_on, source_id
  • clarifications.md
  • proposed-task-changes.md
• Clinic payment reconciliation reviewer: Share total claims, payment records and bank receipts. Find gaps between amounts billed, paid and received.
  • The closing claims receivable, rolled forward from the opening balance
  • Expected bank cash, the actual deposit and the difference between them
  • Every gap named, with what explains it or a note that it is unexplained
  • A list of sources with dates for each figure
  • An editable workbook, if your AI can make spreadsheet files
• me.md builder: Answer a few questions, or let your AI use what it remembers. Get one private file that tells any AI who you are.
  • One me.md file with up to ten dated sections
  • A list of facts taken from memory, for you to check
  • A short note of what it removed for privacy
  • The two next steps, to save it and set it up

Tasks you can use it for

• A referral coordinator is leaving or going on leave, and a colleague takes over the list.
• A referral shows "sent" but the receiving clinic never confirmed it.
• You want to know which referrals are still waiting for a booking date.
• It is review day and you want what changed since last time.

Limits to check before starting

• A patient's condition or urgency is in question. Go to the qualified clinic contact. This agent does not triage.
• You want one quick note for the next administrator and no tracking. Use the Handover note writer.
• You want to send a message or make a booking. This agent only drafts, and you send.
• You want the actions from a team meeting turned into a list. Use the Action item tracker keeper.

Best way to use this role

Give it one clear outcome, the actual files and a definition of done. Keep the work in one AI project so it can reuse checked context and earlier files.

• Add the agent to Claude, ChatGPT or Gemini with its helpers, the Prepare a referral administration handover workflow and the Executive action register keeper. Or download the pack and attach it.
• Remove names, dates of birth and clinical details from your records first. Keep the reference, status and dates. Paste them with your follow-up rules. From the second review, add last review's record.
• Read the list of missing confirmations first. Check that no referral is shown as received or booked without proof.
• Read the draft follow-ups. Edit them, then send them yourself.
• Take anything about a patient's needs to the clinic contact, not to this agent.
• In a chat-only tool, copy the record it gives you and paste it in next time. Installing sets up no schedule, so you start each run.

Read it as data

Read the complete role setup as JSON. Connect your AI to the AI Blocks MCP library and read aiblocks://roles/referral-administration-evidence-review-assistant for its role instructions and pinned sources.

This is a prepared AI role. No task has run. Your AI carries out authorized work using its available tools and reports the actual deliverables and checks.