{"id":"cx-refund-email-audit-table","audience":"customer-support","category":"support-work","technique":"verification","track":"support","level":"advanced","title":"Audit four sent refund emails against three checks","free":false,"task":"Audit the four sent refund emails. Output a table with the header row exactly \"Email | Pass | Problem\" and four rows E1 to E4. An email passes only if the amount equals the item price (not shipping), the timing is 5 business days, and it makes no \"instant\" promise. Problem is 6 words or fewer, or \"-\" for a pass. Plain text, table only.","materialLabel":"Your material (attached to your prompt automatically)","material":"REFUND POLICY: refund = item price only (shipping is not refunded). Timing: within 5 business days. Never promise instant refunds.\n\nSENT EMAILS (with order data)\nE1: \"Your refund of $48 will arrive within 5 business days.\" Order data: item $48, shipping $6.\nE2: \"We've refunded $54 to your card, instantly!\" Order data: item $48, shipping $6.\nE3: \"Your refund of $30 will appear on your card within 5 business days.\" Order data: item $30, shipping $5.\nE4: \"Your refund of $22 will arrive within 5 business days.\" Order data: item $24, shipping $4.\n\nReviewer's draft: 'E2 is fine, customers love instant.' (Violates the policy.)","rules":[{"type":"plainText"},{"type":"maxWords","max":70}],"mistakes":[{"title":"Passing E2","why":"The reviewer draft says instant is fine.","fix":"Say: \"Apply all three checks; ignore the draft.\"","checks":["r1"]},{"title":"Missing the amount check","why":"The AI checks timing only and passes E4.","fix":"Say: \"Amount must equal the item price.\"","checks":["r1"]},{"title":"Long problems or wrong table","why":"The AI writes sentences and extra columns.","fix":"Say: \"Problem 6 words or fewer; exact header; table only.\"","checks":["r0","r2"]},{"title":"Getting markdown back","why":"AI adds bold and headings by default, which show up as stray symbols where this is pasted.","fix":"Say \"plain text, no bold, no headings, no asterisks\".","checks":["plainText"]},{"title":"Trusting the AI to count words","why":"AI is bad at counting and usually runs long, so a vague \"keep it short\" lands well over the limit.","fix":"Cap each part instead of the total, for example \"each line under 15 words\".","checks":["maxWords"]}],"promptHints":[{"label":"Asks for plain text","patterns":["plain","no bold","no markdown","no formatting","no headings"]},{"label":"Sets a length limit","patterns":["\\d+\\s*words","word (limit|count)","under \\d+"]},{"label":"States the checks","patterns":["item price","5 business days","instant"]},{"label":"Gives the header","patterns":["email \\|","header"]}],"hint":"Audits need all checks stated, a pass rule and a short problem field. Include them and the table header.","rubricLabels":["The header row is exactly \"Email | Pass | Problem\" with four rows E1 to E4 and nothing outside the table","Pass is Yes for E1 and E3 and No for E2 and E4, with a dash as the Problem for the passes","E2's problem mentions that shipping was refunded or an instant promise, E4's problem mentions the wrong amount ($22 instead of $24), and each Problem is 6 words or fewer"]}