{"id":"cx-qa-scorecard-reply-evidence","audience":"customer-support","category":"support-work","technique":"verification","track":"support","level":"advanced","title":"Score a support reply on a four-point QA scorecard","free":false,"task":"Score the agent reply on the four criteria below. Output a table with the header \"Criterion | Score | Evidence\", one row per criterion, Score 1 or 0, Evidence a direct quote of fewer than 8 words (or \"none\" if it fails). Then a last line \"Total: N of 4\". Plain text, under 110 words. Judge only what the reply says.","materialLabel":"Your material (attached to your prompt automatically)","material":"CRITERIA\n1. Names the customer\n2. States the current policy correctly (current return window is 30 days)\n3. Gives a next step with a date or time\n4. No blame on the customer\n\nCUSTOMER MESSAGE (Ines): \"I bought a jacket Oct 2 and want to return it. It's Nov 5.\"\n\nAGENT REPLY:\n\"Hi Ines, thanks for writing in. Our return window is 45 days, so you are fine. I've emailed you a prepaid label. Please drop it off by Nov 12 and your refund will be issued 3 to 5 days after we receive it. If you had read the product page you would have seen the fit guide, but no worries.\"\n\nNote: the 45-day window was the policy before this year.","rules":[{"type":"plainText"},{"type":"maxWords","max":110}],"mistakes":[{"title":"Trusting the confident tone","why":"The reply sounds helpful, so the AI gives a pass on policy.","fix":"Say: \"The current window is 30 days. Check every claim against it.\"","checks":["r1"]},{"title":"Missing the sneaky blame","why":"The AI treats the last sentence as friendly.","fix":"Ask: \"Quote the exact phrase for every fail.\"","checks":["r1"]},{"title":"Writing long evidence","why":"The AI paraphrases instead of quoting short evidence.","fix":"Say: \"Evidence: a direct quote under 8 words.\"","checks":["r2"]},{"title":"Free-form scoring","why":"The AI writes paragraphs of feedback instead of a table and total.","fix":"Give the exact header and the Total line format.","checks":["r0"]},{"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","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":"Gives the table header","patterns":["criterion \\|","header","table"]},{"label":"Asks for quotes","patterns":["quote","evidence"]}],"hint":"A scorecard needs the criteria, the real policy, a fixed table and a short quote as proof for each score.","rubricLabels":["The header is \"Criterion | Score | Evidence\", with four rows and Score 1 or 0 on each row","Criterion 1 scores 1, criterion 2 scores 0 (the reply says 45 days), criterion 3 scores 1 (drop off by Nov 12), and criterion 4 scores 0 (the line about reading the product page)","The last line is exactly \"Total: 2 of 4\", and each evidence cell quotes fewer than 8 words from the reply"]}