Customer Email Drafting Agent With Policy Lookup and Escalation Rules
Specify a support drafting agent that checks policy before it writes.
Added October 15, 2026. Written with AI assistance and checked for clarity and placeholders; not reviewed by a lawyer, accountant, clinician or other professional. How these prompts are made
The Open buttons send the prompt text as currently shown below to that site only when you click. Fill in the placeholders first.
Specify a customer-email drafting agent for [BUSINESS_AND_PRODUCT]. Policies and documents it may consult: [POLICY_DOCUMENTS]. Channels: [CHANNELS]. Tone of voice: [TONE]. Who reviews drafts: [REVIEWER_ROLE]. Describe the pipeline: classify the email (question, complaint, refund, bug, billing, sales, abuse), look up the relevant policy section, quote the section it relied on, draft the reply, score its own confidence, and route to a human when confidence is low or the case matches an escalation rule. Provide the escalation rules as a table: trigger (legal threat, safety issue, vulnerable customer, amount above [REFUND_LIMIT], repeated contact, press), what the agent does, and who is notified. Give the format of the draft package: classification, policy quoted, draft, open questions, and a line 'cannot confirm'. Add five example emails with expected classification and action, including one the agent should refuse to answer and hand over. Add rules so it never promises refunds, dates or features that policy does not allow, and never invents order details. Keep the whole specification under 600 words.
Fill in before use
Optional: type your details and the prompt above updates. Nothing is saved or sent.
Replace each bracketed placeholder with your own details, then paste into Any AI or another assistant.