{"id":"eval-mkt-grade-ai-social-captions","audience":"marketing","category":"evaluation","track":"evals","technique":"format","title":"Grade 4 AI social captions with quoted evidence","free":false,"task":"You're spot-checking this week's AI-written Instagram captions. Since yesterday the campaign hashtag is #FernwayFall, not #CozyFall as the brief says, so the old tag now fails C1. Grade the 4 captions in exactly 4 numbered lines: '1. PASS' or '1. FAIL - C2 - \"exact words from the caption\"', naming the first criterion it fails. Plain text, under 75 words.","materialLabel":"The criteria and 4 AI captions (attached to your prompt automatically)","material":"Fernway Outdoor - Instagram caption QA (fall campaign, week 2)\n\nCampaign hashtag (from the Oct brief): #CozyFall\n\nCRITERIA\nC1: Uses the current campaign hashtag.\nC2: Names the featured product exactly as it's listed in the catalog.\nC3: No more than 2 hashtags in total.\n\nCatalog names this week: Ridge Fleece Jacket, Trail Mug 12oz, Summit Daypack 20L, Basecamp Wool Socks\n\nCAPTION 1 (Ridge Fleece Jacket)\n\"First frost? The Ridge Fleece Jacket has you covered. #FernwayFall\"\n\nCAPTION 2 (Trail Mug 12oz)\n\"Hot cocoa, cold trail. Meet the Trail Mug 12oz. #CozyFall\"\n\nCAPTION 3 (Summit Daypack 20L)\n\"Pack light, go far with the Summit Day Pack. #FernwayFall #hiking\"\n\nCAPTION 4 (Basecamp Wool Socks)\n\"Warm toes, happy hikes. Basecamp Wool Socks are back. #FernwayFall #hiking #fallvibes\"\n\nNotes\n- Last campaign, two different hashtags split our reach.\n- The social lead wants grades she can paste into the tracker, one line each.","rules":[{"type":"numberedItems","count":4},{"type":"plainText"},{"type":"maxWords","max":75}],"mistakes":[{"title":"Missing the hashtag change","why":"The AI doesn't know the tag changed yesterday, so it passes #CozyFall.","fix":"Tell it: \"The current hashtag is #FernwayFall. #CozyFall fails C1.\"","checks":["r0"]},{"title":"No evidence behind a FAIL","why":"The AI writes 'wrong product name' without showing the words.","fix":"Say: \"Quote the exact words from the caption that fail it.\"","checks":["r1"]},{"title":"Rewrites instead of grades","why":"The AI 'helps' by rewriting each caption.","fix":"Show the shape: \"3. FAIL - C2 - \\\"quote\\\"\", no rewrites.","checks":["r2","numberedItems"]},{"title":"Getting markdown back","why":"AI defaults to bold labels and a table that won't paste into the tracker.","fix":"Ask for \"plain text, no bold, no tables.\"","checks":["plainText","maxWords"]}],"promptHints":[{"label":"Shows the line shape","patterns":["FAIL - C","PASS","one line per"]},{"label":"Asks for a quote as evidence","patterns":["quote","exact words","evidence"]},{"label":"Gives the hashtag change","patterns":["fernwayfall","cozyfall","hashtag"]},{"label":"Says exactly how many","patterns":["\\b(4|four)\\b.{0,25}(line|caption)","exactly \\d+"]},{"label":"Asks for plain text","patterns":["plain","no bold","no markdown","no formatting","no headings"]}],"hint":"Spot-checks only work if every grade can be checked. Show the AI the one-line shape, ask for the quote behind each fail, and tell it what changed this week.","rubricLabels":["Grades Caption 1 PASS, Caption 2 FAIL on C1 (#CozyFall is the old tag), Caption 3 FAIL on C2 (\"Day Pack\" instead of Daypack 20L), and Caption 4 FAIL on C3 (three hashtags)","Every FAIL line quotes exact words from that caption in quotation marks (for example \"#CozyFall\", \"Summit Day Pack\" or the three hashtags)","Each line uses only the shape 'N. PASS' or 'N. FAIL - Cx - \"quote\"' with no extra commentary, suggested rewrites, headings or summary"]}