{"id":"risk-vendor-tier-buyer-faq","audience":"compliance","category":"legal-compliance","track":"legal","technique":"audience","level":"intermediate","title":"Write a four-question FAQ for buyers on vendor tiering","free":false,"task":"Write a prompt for an FAQ aimed at buyers who are not risk specialists. Plain text, exactly four pairs labeled Q1:/A1: through Q4:/A4:, each answer one sentence under 25 words, under 170 words in total. Use only the standard. One question must cover what a buyer does when they are unsure of a vendor's tier. Do not interpret any law.","materialLabel":"Your material (attached to your prompt automatically)","material":"VENDOR RISK STANDARD v1 (summary for buyers)\nTiers: Tier 1 = personal data on more than 10,000 people, or can initiate payments or refunds. Tier 2 = personal data on 10,000 people or fewer, cannot move money. Tier 3 = no personal data.\nReviews: Tier 1 vendors are reviewed every year. Tier 2 every two years. Tier 3 are not reviewed unless something changes.\nWho decides: the Vendor Risk Owner (currently Elena Marsh) confirms every tier. Buyers propose using the intake form.\nIf data volumes change by more than 10%, the buyer tells the Vendor Risk Owner.\nThe standard is a company document. It does not describe what any law requires.","rules":[{"type":"plainText"},{"type":"maxWords","max":170}],"promptHints":[{"label":"Q/A labels","patterns":["q1:","a1:","labeled"]},{"label":"Four pairs","patterns":["four","exactly 4"]},{"label":"Unsure case","patterns":["unsure","risk owner"]},{"label":"One sentence","patterns":["one sentence","under \\d+"]},{"label":"Standard only","patterns":["only the standard","use only"]}],"mistakes":[{"title":"Missing the unsure question","why":"AI covers easy topics and skips what to do when unsure.","fix":"Say: \"One question covers what a buyer does when unsure of a tier.\"","checks":["r0"]},{"title":"Long answers","why":"AI writes paragraphs per answer.","fix":"Say: \"Each answer one sentence under 22 words.\"","checks":["r1"]},{"title":"Wrong labels","why":"AI uses bold questions or numbering without Q/A labels.","fix":"Say: \"Labels exactly Q1: A1: ... Q4: A4:, plain text.\"","checks":["r2"]},{"title":"Getting markdown back","why":"AI adds bold, headings and bullets by default, which show up as stray symbols when 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 over the limit.","fix":"Cap each part instead of the total, for example \"each line under 15 words\".","checks":["maxWords"]}],"hint":"Give the Q/A labels, the topics to cover and the one-sentence limit.","mission":"Answer buyers' tiering questions in four pairs.","musts":["Four labeled pairs","Unsure case covered","Standard only"],"rubricLabels":["There are exactly four pairs labeled Q1:/A1: through Q4:/A4: in plain text, each answer a single sentence","One answer says that when unsure of the tier the buyer asks the Vendor Risk Owner (Elena Marsh) or submits the intake form for confirmation","Answers use the review cycles and thresholds from the standard (yearly for Tier 1, every two years for Tier 2, 10,000 people, 10% change) and none interprets a law"]}