{"id":"acct-credit-limit-by-policy-three-tests-lines","audience":"accounting","category":"finance-work","track":"finance","technique":"constraints","level":"advanced","title":"Size a credit limit by a lowest-of-three policy","free":false,"task":"You're helping the credit manager size a first credit limit. Write a prompt for plain text with lines starting \"Net worth used:\", \"Test A:\", \"Test B:\", \"Test C:\", \"Limit by policy:\" and \"Flag:\". Use the figure the policy names when sources disagree. Limit by policy is the lowest of the three tests. Show the math. The Flag line notes anything for the credit manager to review. Do not approve the limit. Under 120 words.","materialLabel":"Source material (attached to your prompt automatically)","material":"CREDIT APPLICATION - Lakeshore Print (new customer)\n\nPolicy for first credit limits - the limit is the LOWEST of:\n- Test A: 10% of tangible net worth shown on the applicant's most recent financial statements\n- Test B: 2 times the applicant's largest prior single order with us; if there is no prior order, $10,000\n- Test C: $50,000, the cap for any first credit limit\n\nApplicant's data:\n- Net worth on the application form (self-reported): $2,000,000\n- Tangible net worth on the financial statements dated Jun 30, 2026: $1,200,000\n- Prior orders with us: none\n- Trade references: two of three returned; one reference reports the applicant pays about 25 days late on average\n\nNotes: the sales rep says the form figure is \"more recent, so use it.\" The policy names the financial statements. Credit limits are approved by the credit manager, Samira Khan, not by analysts. The first order is expected to be $18,000.","rules":[{"type":"plainText"},{"type":"maxWords","max":120}],"promptHints":[{"label":"Names the six lines","patterns":["net worth used","test a","test b","test c","limit by policy","flag"]},{"label":"Uses the policy source","patterns":["financial statements","policy names"]},{"label":"Takes the lowest","patterns":["lowest","lesser"]},{"label":"Bans approval","patterns":["do not approve","credit manager"]},{"label":"Caps length","patterns":["\\d+\\s*words","under \\d+"]}],"mistakes":[{"title":"Using the application's net worth","why":"The AI uses $2,000,000 because it is newer.","fix":"Say: \"Use the financial statements the policy names.\"","checks":["r0"]},{"title":"Choosing the wrong test","why":"The AI takes Test A's $120,000 or the $50,000 cap.","fix":"Add: \"The limit is the lowest of the three tests.\"","checks":["r1"]},{"title":"Approving the limit","why":"The AI says the limit is approved.","fix":"Add: \"Do not approve; the credit manager decides.\"","checks":["r2"]},{"title":"Missing flags","why":"The AI doesn't note the late reference or the order size.","fix":"Add: \"Flag: anything for the credit manager to review.\"","checks":["r2"]},{"title":"Markdown extras","why":"The AI adds bold labels, headings and a closing note.","fix":"Say: \"Plain text, no bold. Output only the lines asked for.\"","checks":["plainText"]},{"title":"Trusting the AI to count words","why":"AI is bad at counting and runs past word limits.","fix":"Ask for fewer words than the real cap and keep each part short.","checks":["maxWords"]}],"hint":"The policy names the source and the formula. State both, then ask for flags without approval.","rubricLabels":["Net worth used shows $1,200,000 from the financial statements (not the $2,000,000 on the form) and the six lines start with the required labels in order","Test A shows 10% x $1,200,000 = $120,000, Test B shows $10,000 (no prior order), Test C shows $50,000, and Limit by policy shows the lowest, $10,000","Flag mentions the late-paying reference (about 25 days late) and that the expected $18,000 first order is above the $10,000 limit, and the output does not approve the limit"]}