Design a Customer-Enquiry Auto-Responder (Draft-for-Approval)
Set up an AI that drafts fast, on-brand replies to common enquiries for you to approve and send.
When to use it: When enquiry emails pile up and you want AI to draft accurate replies you quickly check before sending.
You are an automation designer helping an Australian small business handle repetitive customer enquiries with AI — safely. The AI drafts; a human approves and sends. It never auto-sends, never quotes a price or commitment the owner hasn't confirmed, and never invents availability. Your job is to design the drafting system and the reply templates.
<context>
[BUSINESS]: what you do and where.
[COMMON ENQUIRIES]: the questions you get most (e.g. 'do you service my suburb?', 'how much for X?', 'are you available Saturday?').
[FACTS THE AI CAN USE]: confirmed info the AI may state — service areas, hours, standard inclusions, price ranges you're happy to publish, booking process. Only what's true and approved.
[TONE]: how you want to sound (e.g. 'friendly, straight-talking, no jargon').
[ESCALATE WHEN]: enquiries that must come straight to you untouched (complaints, custom jobs, anything unusual).
</context>
<task>
First, group [COMMON ENQUIRIES] into types and flag which can be safely drafted vs which must escalate under [ESCALATE WHEN].
Then produce:
1. A system instruction for the AI: its role, the approved facts it may state, the tone, the hard rule that it drafts only (a human sends), and the exact escalation triggers.
2. A reply-draft template for each safe enquiry type, using only [FACTS THE AI CAN USE], with [PLACEHOLDERS] where the owner must confirm a specific (a date, a firm price).
3. A one-line 'why this is safe' note per template (what it will and won't commit to).
</task>
<output_format>
- Enquiry types: draft-safe vs escalate
- The AI system instruction (ready to use), with the draft-only + escalation rules explicit
- One reply template per safe enquiry type, with [PLACEHOLDERS] for owner confirmation
- A short 'go-live checklist' (approve first 10 drafts before trusting the pattern)
en-AU spelling, warm and clear.
</output_format>
Grounding: use only [FACTS THE AI CAN USE]. Never let the AI state prices, availability, timeframes or guarantees that weren't provided — those become [PLACEHOLDER] for the owner. The system must draft for human approval, never auto-send. Mark gaps as [NEEDED: ...].
Copy the block above straight into Claude — anything in [BRACKETS] is yours to fill in.
Want it tuned to your business? Bring it to the free weekly call and we'll adapt it live.
Join the free callMore ai agents & automation prompts
Map Your Business for AI Automation
Find the repetitive, rule-based tasks in your week that are actually worth handing to AI — and the ones to leave alone.
Write a Clear Brief for an AI Agent to Do a Task
Turn a fuzzy 'can AI do this?' into a precise, testable brief an AI can follow reliably.
Build a Standard Operating Procedure an AI Can Follow
Document a recurring task as a clear SOP that both a new staff member and an AI can execute consistently.