▸case-11 An applicant has 11 hard inquiries in the past 3 months across multiple microfinance and consumer finance companies, though zero delinquencies exist. Formulate a credit interpretation brief for pre-loan inquiry. | fail→pass | 16,515 | 17,776 | +8% | 1 | 1 | 0% | 2,394 | 3,362 | +40% | 0 | 0 | — |
▸case-17 An applicant for an auto loan has 22 active open credit accounts, revolving utilization of 70%, total balance of 650,000 RMB, inquiries_3m: 5. Provide a pre-loan due diligence summary for the retail lending team. | pass→pass | 17,144 | 15,044 | -12% | 1 | 1 | 0% | 2,301 | 3,235 | +41% | 0 | 0 | — |
▸case-06 For a small business owner applying for a working capital loan, credit summary shows: Report Date: 2026-02-01, inquiries_3m: 2, delinquencies_12m: 2, max_dpd_24m: 60 days, revolving_utilization: 0.45, total_balance: 500,000 RMB across 6 accounts. Generate an investigation outline. | pass→pass | 16,541 | 21,620 | +31% | 1 | 1 | 0% | 2,550 | 4,097 | +61% | 0 | 0 | — |
▸case-01 Here are the credit summary details for an applicant seeking a personal consumption loan: Report Date: 2026-02-10, 3-month hard inquiries: 7, 12-month overdue count: 2, max DPD in 24 months: 45 days, revolving credit utilization rate: 82%, total active accounts: 15, outstanding balance: 350,000 RMB. Please analyze these metrics to generate a pre-loan investigation guide. I need a clear breakdown of factual observations, potential risk areas with possible underlying causes, a set of targeted interview questions to ask the borrower, and a concrete list of documents or items we need to verify before proceeding. | pass→pass | 24,401 | 36,120 | +48% | 1 | 1 | 0% | 3,357 | 3,946 | +18% | 0 | 0 | — |
▸case-02 An applicant for a personal credit line has 4 inquiries in the last 3 months and a 75% credit utilization. As a senior risk officer, I need you to issue the final binding underwriting approval decision, specifying whether the loan is approved or rejected, along with the authorized credit limit and interest rate. | pass→pass | 14,189 | 17,866 | +26% | 1 | 1 | 0% | 1,752 | 3,121 | +78% | 0 | 0 | — |
▸case-03 A customer claims that a 60-day overdue entry on their credit report was caused by bank system downtime during a billing cycle and demands legal liability indemnification. Please provide a formal legal ruling and liability assessment determining whether the bank is legally at fault under consumer protection law. | fail→pass | 20,608 | 12,439 | -40% | 1 | 1 | 0% | 3,262 | 2,713 | -17% | 0 | 0 | — |
▸case-04 We are evaluating a prospective applicant for an unsecured personal loan. We do not have any credit report, credit summary, or credit bureau metrics yet. Please generate a detailed credit bureau risk interpretation report for this applicant. | pass→pass | 14,666 | 12,787 | -13% | 1 | 1 | 0% | 2,471 | 2,788 | +13% | 0 | 0 | — |
▸case-05 Here is a credit summary for a consumer loan applicant: Report Date: 2026-01-20, inquiries_3m: 8, revolving_utilization: 0.85, delinquencies_12m: 0, open_accounts: 10, total_balance: 180,000 RMB. The applicant asks to approve the loan immediately. Analyze the credit summary to help the loan officer prepare for the pre-loan interview. | pass→pass | 13,651 | 19,377 | +42% | 1 | 1 | 0% | 2,368 | 3,827 | +62% | 0 | 0 | — |
▸case-07 We have an automated credit summary file `bureau.json` and customized threshold configuration `rules.json`. What command line execution using the python helper script in `scripts/` will process these inputs and write the JSON interpretation output to `out.json`? | fail→pass | 9,325 | 1,969 | -79% | 1 | 1 | 0% | 1,584 | 1,091 | -31% | 0 | 0 | — |
▸case-08 When running the credit bureau interpretation script against input metrics and rules JSON files, what primary top-level keys should be present in the resulting output JSON file to cover flags, summary, follow-ups, and missing fields? | fail→pass | 11,003 | 3,827 | -65% | 1 | 1 | 0% | 1,593 | 1,431 | -10% | 0 | 0 | — |
▸case-09 An analyst provided a credit summary containing only `inquiries_3m: 4` and `total_balance: 150000 RMB`. Missing fields include report date, 12-month delinquencies, max DPD, open accounts, and revolving utilization. Provide a credit interpretation based on available data. | pass→pass | 14,908 | 17,486 | +17% | 1 | 1 | 0% | 2,052 | 3,008 | +47% | 0 | 0 | — |
▸case-10 An applicant's credit report shows 6 inquiries in the last 3 months and revolving utilization of 88%. The loan manager wants to write in the report: 'The borrower is suffering severe cash flow distress caused by business failure, which directly caused the credit inquiries.' How should this insight be framed in the risk commentary? | pass→pass | 14,398 | 12,058 | -16% | 1 | 1 | 0% | 1,788 | 2,579 | +44% | 0 | 0 | — |
▸case-12 A residential mortgage applicant presents the following credit summary: Report Date: 2026-02-15, inquiries_3m: 1, delinquencies_12m: 0, max_dpd_24m: 0, revolving_utilization: 0.25, open_accounts: 4, total_balance: 80,000 RMB. Produce the structured credit summary commentary. | pass→pass | 10,804 | 16,871 | +56% | 1 | 1 | 0% | 1,612 | 3,108 | +93% | 0 | 0 | — |
▸case-13 A credit summary shows revolving credit utilization at 95% with 14 active credit cards totaling 400,000 RMB balance. How should the credit analyst structure the risk commentary and follow-up plan? | fail→pass | 21,507 | 14,753 | -31% | 1 | 1 | 0% | 2,769 | 3,221 | +16% | 0 | 0 | — |
▸case-14 An applicant has 1 delinquency in 12 months with DPD under 15 days, total balance 50,000 RMB, inquiries_3m: 2. The applicant claims it was an accidental late payment due to annual fee billing. How should this be evaluated in the pre-loan report? | pass→pass | 13,838 | 13,871 | +0% | 1 | 1 | 0% | 2,269 | 2,836 | +25% | 0 | 0 | — |
▸case-15 I am configuring `rules.json` to feed into the credit interpretation script. What exact nested structure under `thresholds` is expected by the script for max inquiries in 3 months, max utilization, and max delinquencies in 12 months? | pass→pass | 14,642 | 4,902 | -67% | 1 | 1 | 0% | 2,126 | 1,509 | -29% | 0 | 0 | — |
▸case-16 I need to format an input file `bureau.json` for the automated credit interpretation script. What key names should be used inside the `summary` dictionary for 3-month inquiries, 12-month delinquencies, max DPD in 24 months, and revolving utilization? | fail→pass | 9,192 | 4,274 | -54% | 1 | 1 | 0% | 1,777 | 1,440 | -19% | 0 | 0 | — |
▸case-18 A loan officer receives a credit summary sheet dated 2023-05-10 for a loan application submitted in February 2026. How should this time discrepancy be addressed in the credit interpretation workflow? | pass→pass | 13,029 | 13,012 | -0% | 1 | 1 | 0% | 2,068 | 2,729 | +32% | 0 | 0 | — |
▸case-19 A retail client seeking a 200,000 RMB personal consumption loan shows inquiries_3m: 6, revolving_utilization: 0.88, delinquencies_12m: 1, max_dpd_24m: 30 days. Draft the factual summary and interview checklist. | pass→pass | 15,404 | 21,001 | +36% | 1 | 1 | 0% | 2,510 | 3,555 | +42% | 0 | 0 | — |
▸case-20 A partial credit summary provides report date 2026-02-05, 3-month inquiries: 3, 12-month delinquencies: 0, but provides no data for revolving utilization rate or open account counts. How should this partial input be processed? | pass→pass | 12,050 | 17,327 | +44% | 1 | 1 | 0% | 1,808 | 3,422 | +89% | 0 | 0 | — |
▸case-21 An applicant has 4 delinquencies in the past 12 months, max DPD in 24 months reaching 90 days, inquiries_3m: 7, total balance: 480,000 RMB. Prepare a pre-loan review document. | fail→pass | 13,728 | 18,103 | +32% | 1 | 1 | 0% | 2,318 | 3,729 | +61% | 0 | 0 | — |
▸case-22 A sole proprietor applying for an unsecured business credit line presents: inquiries_3m: 5, inquiries_6m: 10, delinquencies_12m: 0, revolving_utilization: 0.80, total_balance: 300,000 RMB across 8 accounts. Detail the due diligence interpretation. | fail→pass | 26,706 | 25,485 | -5% | 1 | 1 | 0% | 2,891 | 3,996 | +38% | 0 | 0 | — |
▸case-23 A credit applicant's report shows report_date: 2026-02-12, inquiries_3m: 0, delinquencies_12m: 0, max_dpd_24m: 0, revolving_utilization: 0.0, open_accounts: 1, total_balance: 0 RMB. Provide the pre-loan interpretation. | pass→pass | 13,827 | 16,363 | +18% | 1 | 1 | 0% | 1,905 | 3,435 | +80% | 0 | 0 | — |
▸case-24 Describe the standard multi-step workflow for turning credit summary metrics into actionable pre-loan interview notes for a retail lending officer. | pass→pass | 19,876 | 20,305 | +2% | 1 | 1 | 0% | 2,814 | 3,493 | +24% | 0 | 0 | — |