Support Reply Accuracy and Policy Fit Audit
Check replies against policy and facts.
The Open buttons send the prompt text as currently shown below to that site only when you click. Fill in the placeholders first.
[VERIFY] is an editorial flag, not a blank to fill. Wherever the prompt shows [VERIFY], check that item against a reliable source before you use it.
Audit replies. Ten replies with customer messages and the relevant policy text: [REPLIES_AND_POLICY]. Check each statement against the policy, flag promises beyond policy, wrong facts and missing next steps, and score accuracy from 1 to 5. Return a table, a corrected reply for the weakest and a coaching note. Add a worked scoring on an invented reply: customer asks about a refund after 35 days when the policy allows 30; reply says No refunds after 30 days, accuracy 2 because it ignores a faulty-item exception, tone 2, next step 1, total 5 out of 15. Better: I'm sorry it has not worked out. Our 30-day window has passed, but if the item is faulty we can look at a repair or refund; please send a photo and the order number. Include a check against statutory rights [VERIFY]. Add a worked scoring on a sample reply: accuracy 2 because the reply promised a refund the policy does not allow; completeness 3; empathy 4; next step 1. Total 10 out of 20. Provide a corrected reply under 100 words and a coaching note that explains the policy gently.
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.