Let an agent fully answer routine customer emails, with clear rules for when to hand off: AI agent prompt
Instead of an AI that only drafts replies for staff to approve, define which routine email requests the agent may resolve end to end (missing card, payment questions, status updates) and exactly when it must stop and escalate to a person, such as complaints, vulnerable customers or anything outside approved steps. UK digital bank LHV Bank just put a Gradient Labs agent live on its customer email support this way, with every change tested and approved before release.
Customer SupportEmail support automationintermediateThe prompt
You are helping me design an email support agent for my business. Business: [type, what customers usually email about]. Our policies and procedures: [paste refund, account, shipping or other rules]. Tools the agent could use: [e.g. order lookup, ticket status, help center]. Step 1: from these sample emails [paste 10-20 recent anonymized emails], group requests into types and count them. Step 2: mark each type as 'agent can resolve end to end', 'agent drafts, human approves' or 'always escalate to a human', and explain why using only our policies. Step 3: write the escalation rules as a clear checklist (complaints, upset or vulnerable customers, money above [amount], anything not covered by policy). Step 4: write reply templates for the top 3 'resolve' types. Step 5: list 10 test emails, including tricky ones, to check the agent before going live. Do not invent policies; mark gaps as 'needs a decision'.
How to use it
- Copy the prompt above.
- Replace every [bracket] with your own details (or fill in the blanks on AgentPlaybook Real Estate, which remembers your answers for the next prompt).
- Paste it into Gradient Labs, Zendesk, Gmail and review the answer before you use it.
Works with: Gradient Labs, Zendesk, Gmail, ChatGPT, Claude
Real-world example: fintech.global/2026/10/21/lhv-bank-launches-gradient-labs-ai-agent-for
Featured in: ChatGPT prompts for business email, replies and follow-ups · ChatGPT prompts for ecommerce and online stores (Shopify, Etsy, WooCommerce) · ChatGPT prompts for customer service: ticket triage, replies and handoffs
Before you use it: what to check
Already built into this prompt
- Do not invent policies; mark gaps as 'needs a decision'.
Check yourself
- This prompt works on text other people wrote (emails, reviews, messages or web pages). That text can hide instructions aimed at AI, such as "ignore your rules and…". Paste it as material to work on, and never let the AI act on requests inside it, like paying, sharing logins or changing records. How prompt injection works →
- Check every figure, price and total against your own records before anything is sent, posted or booked. A person approves refunds, discounts, payments and price changes.
- Remove names, contact details, account numbers and other personal data the task doesn't need before you paste, and check your AI tool's data and training settings.
- Read and edit the draft before it reaches a customer or goes public. If you later automate this prompt, start with "draft and notify me", not "send automatically".
Using an AI agent? Download this prompt as a SKILL.md agent skill with these checks built in, or get the .zip for Claude.ai. How to install skills.
Want to make this prompt even better? Grade it with the free prompt grader or build your own with the AI prompt generator.
32 copy-paste agent prompts (PDF) plus 5 new prompts every week.
Get it freeMore Customer Support prompts
See all Customer Support prompts →