Understand the failure.
Verify the repair.
Small, reproducible software failures. The broken implementation, the fix that didn’t work, and the one that passed—preserved together.
Explore the cases ↓How results are verified ↗WHAT THE ARCHIVE CONTAINS
100840 executable cases. 20168 are open.
Every case records the implementation that fails, the fix that did not work, and the repair that passed its checks—with recorded outputs and source hashes. This release adds 100840 cases across 20168 failure mechanisms and 254 domains.
The open tier gives you the failure and the unsuccessful fix for one case in every mechanism. The remaining 80672 cases, 5 variants per mechanism, are member-only: the verified repair, its recorded checks, and the full fixture suite are held in the member archive. Read the methodology ↗
A RECORD OF WHAT WENT WRONG
Browse the archive / 100840
mcg/kg/min infusion converted to pump mL/h: per-minute to per-hour factor · case 01
Programmed rates are one sixtieth of the intended rate.
mcg/kg/min infusion converted to pump mL/h: per-minute to per-hour factor · case 02
Programmed rates are one sixtieth of the intended rate.
mcg/kg/min infusion converted to pump mL/h: per-minute to per-hour factor · case 03
Programmed rates are one sixtieth of the intended rate.
mcg/kg/min infusion converted to pump mL/h: per-minute to per-hour factor · case 04
Programmed rates are one sixtieth of the intended rate.
mcg/kg/min infusion converted to pump mL/h: per-minute to per-hour factor · case 05
Programmed rates are one sixtieth of the intended rate.
mcg/kg/min infusion converted to pump mL/h: rate rounding to pump resolution · case 01
Rates such as 2.25 mL/h are programmed as 2.2 mL/h.
mcg/kg/min infusion converted to pump mL/h: rate rounding to pump resolution · case 02
Rates such as 2.25 mL/h are programmed as 2.2 mL/h.
mcg/kg/min infusion converted to pump mL/h: rate rounding to pump resolution · case 03
Rates such as 2.25 mL/h are programmed as 2.2 mL/h.
mcg/kg/min infusion converted to pump mL/h: rate rounding to pump resolution · case 04
Rates such as 2.25 mL/h are programmed as 2.2 mL/h.
mcg/kg/min infusion converted to pump mL/h: rate rounding to pump resolution · case 05
Rates such as 2.25 mL/h are programmed as 2.2 mL/h.
mcg/kg/min infusion converted to pump mL/h: pump ceiling comparison · case 01
A rate of exactly 999.9 mL/h is reported as exceeding the pump.
mcg/kg/min infusion converted to pump mL/h: pump ceiling comparison · case 02
A rate of exactly 999.9 mL/h is reported as exceeding the pump.
mcg/kg/min infusion converted to pump mL/h: pump ceiling comparison · case 03
A rate of exactly 999.9 mL/h is reported as exceeding the pump.
mcg/kg/min infusion converted to pump mL/h: pump ceiling comparison · case 04
A rate of exactly 999.9 mL/h is reported as exceeding the pump.
mcg/kg/min infusion converted to pump mL/h: pump ceiling comparison · case 05
A rate of exactly 999.9 mL/h is reported as exceeding the pump.
mcg/kg/min infusion converted to pump mL/h: zero-rate resolution guard · case 01
A rate of 0.04 mL/h is returned as OK with a 0.0 mL/h program.
mcg/kg/min infusion converted to pump mL/h: zero-rate resolution guard · case 02
A rate of 0.04 mL/h is returned as OK with a 0.0 mL/h program.
mcg/kg/min infusion converted to pump mL/h: zero-rate resolution guard · case 03
A rate of 0.04 mL/h is returned as OK with a 0.0 mL/h program.
mcg/kg/min infusion converted to pump mL/h: zero-rate resolution guard · case 04
A rate of 0.04 mL/h is returned as OK with a 0.0 mL/h program.
mcg/kg/min infusion converted to pump mL/h: zero-rate resolution guard · case 05
A rate of 0.04 mL/h is returned as OK with a 0.0 mL/h program.
Cockcroft-Gault renal dose adjustment: elderly creatinine floor age gate · case 01
A 65-year-old with low creatinine gets an inflated clearance estimate.
Cockcroft-Gault renal dose adjustment: elderly creatinine floor age gate · case 02
A 65-year-old with low creatinine gets an inflated clearance estimate.
Cockcroft-Gault renal dose adjustment: elderly creatinine floor age gate · case 03
A 65-year-old with low creatinine gets an inflated clearance estimate.
Cockcroft-Gault renal dose adjustment: elderly creatinine floor age gate · case 04
A 65-year-old with low creatinine gets an inflated clearance estimate.
Cockcroft-Gault renal dose adjustment: elderly creatinine floor age gate · case 05
A 65-year-old with low creatinine gets an inflated clearance estimate.
Cockcroft-Gault renal dose adjustment: creatinine floor direction · case 01
Elderly patients with impaired function are assigned a creatinine of 0.8 and look healthy.
Cockcroft-Gault renal dose adjustment: creatinine floor direction · case 02
Elderly patients with impaired function are assigned a creatinine of 0.8 and look healthy.
Cockcroft-Gault renal dose adjustment: creatinine floor direction · case 03
Elderly patients with impaired function are assigned a creatinine of 0.8 and look healthy.
Cockcroft-Gault renal dose adjustment: creatinine floor direction · case 04
Elderly patients with impaired function are assigned a creatinine of 0.8 and look healthy.
Cockcroft-Gault renal dose adjustment: creatinine floor direction · case 05
Elderly patients with impaired function are assigned a creatinine of 0.8 and look healthy.
Cockcroft-Gault renal dose adjustment: clearance truncation · case 01
A clearance of 59.7 mL/min is treated as 60 and the interval is not extended.
Cockcroft-Gault renal dose adjustment: clearance truncation · case 02
A clearance of 59.7 mL/min is treated as 60 and the interval is not extended.
Cockcroft-Gault renal dose adjustment: clearance truncation · case 03
A clearance of 59.7 mL/min is treated as 60 and the interval is not extended.
Cockcroft-Gault renal dose adjustment: clearance truncation · case 04
A clearance of 59.7 mL/min is treated as 60 and the interval is not extended.
Cockcroft-Gault renal dose adjustment: clearance truncation · case 05
A clearance of 59.7 mL/min is treated as 60 and the interval is not extended.
Cockcroft-Gault renal dose adjustment: moderate band lower edge · case 01
A clearance of exactly 30 mL/min receives the severe-band half dose.
Cockcroft-Gault renal dose adjustment: moderate band lower edge · case 02
A clearance of exactly 30 mL/min receives the severe-band half dose.
Cockcroft-Gault renal dose adjustment: moderate band lower edge · case 03
A clearance of exactly 30 mL/min receives the severe-band half dose.
Cockcroft-Gault renal dose adjustment: moderate band lower edge · case 04
A clearance of exactly 30 mL/min receives the severe-band half dose.
Cockcroft-Gault renal dose adjustment: moderate band lower edge · case 05
A clearance of exactly 30 mL/min receives the severe-band half dose.
Cockcroft-Gault renal dose adjustment: severe band dose rounding · case 01
A 175 mg dose is halved to 87 mg, which is not a 5 mg multiple.
Cockcroft-Gault renal dose adjustment: severe band dose rounding · case 02
A 175 mg dose is halved to 87 mg, which is not a 5 mg multiple.
Cockcroft-Gault renal dose adjustment: severe band dose rounding · case 03
A 175 mg dose is halved to 87 mg, which is not a 5 mg multiple.
Cockcroft-Gault renal dose adjustment: severe band dose rounding · case 04
A 175 mg dose is halved to 87 mg, which is not a 5 mg multiple.
Cockcroft-Gault renal dose adjustment: severe band dose rounding · case 05
A 175 mg dose is halved to 87 mg, which is not a 5 mg multiple.
Cockcroft-Gault renal dose adjustment: severe band interval extension · case 01
Severe impairment halves the dose but keeps the original interval.
Cockcroft-Gault renal dose adjustment: severe band interval extension · case 02
Severe impairment halves the dose but keeps the original interval.
Cockcroft-Gault renal dose adjustment: severe band interval extension · case 03
Severe impairment halves the dose but keeps the original interval.
Cockcroft-Gault renal dose adjustment: severe band interval extension · case 04
Severe impairment halves the dose but keeps the original interval.
Cockcroft-Gault renal dose adjustment: severe band interval extension · case 05
Severe impairment halves the dose but keeps the original interval.
Tablet strength selection with scored halves: half-tablet eligibility · case 01
Half of an unscored tablet is dispensed.
Tablet strength selection with scored halves: half-tablet eligibility · case 02
Half of an unscored tablet is dispensed.
Tablet strength selection with scored halves: half-tablet eligibility · case 03
Half of an unscored tablet is dispensed.
Tablet strength selection with scored halves: half-tablet eligibility · case 04
Half of an unscored tablet is dispensed.
Tablet strength selection with scored halves: half-tablet eligibility · case 05
Half of an unscored tablet is dispensed.
Tablet strength selection with scored halves: tablet unit rounding · case 01
A 14 mg order with 5 mg tablets finds no match although 3 tablets are within tolerance.
Tablet strength selection with scored halves: tablet unit rounding · case 02
A 14 mg order with 5 mg tablets finds no match although 3 tablets are within tolerance.
Tablet strength selection with scored halves: tablet unit rounding · case 03
A 14 mg order with 5 mg tablets finds no match although 3 tablets are within tolerance.
Tablet strength selection with scored halves: tablet unit rounding · case 04
A 14 mg order with 5 mg tablets finds no match although 3 tablets are within tolerance.
Tablet strength selection with scored halves: tablet unit rounding · case 05
A 14 mg order with 5 mg tablets finds no match although 3 tablets are within tolerance.
Tablet strength selection with scored halves: maximum tablet count · case 01
A 3.5-tablet regimen of a scored product is rejected.
Tablet strength selection with scored halves: maximum tablet count · case 02
A 3.5-tablet regimen of a scored product is rejected.
Tablet strength selection with scored halves: maximum tablet count · case 03
A 3.5-tablet regimen of a scored product is rejected.
Tablet strength selection with scored halves: maximum tablet count · case 04
A 3.5-tablet regimen of a scored product is rejected.
Tablet strength selection with scored halves: maximum tablet count · case 05
A 3.5-tablet regimen of a scored product is rejected.
Tablet strength selection with scored halves: tolerance percentage base · case 01
Large doses are rejected even when the percentage deviation is within tolerance.
Tablet strength selection with scored halves: tolerance percentage base · case 02
Large doses are rejected even when the percentage deviation is within tolerance.
Tablet strength selection with scored halves: tolerance percentage base · case 03
Large doses are rejected even when the percentage deviation is within tolerance.
Tablet strength selection with scored halves: tolerance percentage base · case 04
Large doses are rejected even when the percentage deviation is within tolerance.
Tablet strength selection with scored halves: tolerance percentage base · case 05
Large doses are rejected even when the percentage deviation is within tolerance.
Tablet strength selection with scored halves: candidate ranking · case 01
Two tablets of a smaller strength are chosen over one and a half of a larger strength giving the same dose.
Tablet strength selection with scored halves: candidate ranking · case 02
Two tablets of a smaller strength are chosen over one and a half of a larger strength giving the same dose.
Tablet strength selection with scored halves: candidate ranking · case 03
Two tablets of a smaller strength are chosen over one and a half of a larger strength giving the same dose.
Tablet strength selection with scored halves: candidate ranking · case 04
Two tablets of a smaller strength are chosen over one and a half of a larger strength giving the same dose.
Tablet strength selection with scored halves: candidate ranking · case 05
Two tablets of a smaller strength are chosen over one and a half of a larger strength giving the same dose.
Meal and correction insulin bolus with insulin on board: hypoglycaemia hold threshold · case 01
A glucose of exactly 70 mg/dL blocks a meal bolus.
Meal and correction insulin bolus with insulin on board: hypoglycaemia hold threshold · case 02
A glucose of exactly 70 mg/dL blocks a meal bolus.
Meal and correction insulin bolus with insulin on board: hypoglycaemia hold threshold · case 03
A glucose of exactly 70 mg/dL blocks a meal bolus.
Meal and correction insulin bolus with insulin on board: hypoglycaemia hold threshold · case 04
A glucose of exactly 70 mg/dL blocks a meal bolus.
Meal and correction insulin bolus with insulin on board: hypoglycaemia hold threshold · case 05
A glucose of exactly 70 mg/dL blocks a meal bolus.
Meal and correction insulin bolus with insulin on board: insulin on board scope · case 01
Insulin on board keeps eating into the meal dose after the correction reaches zero.
Meal and correction insulin bolus with insulin on board: insulin on board scope · case 02
Insulin on board keeps eating into the meal dose after the correction reaches zero.
Meal and correction insulin bolus with insulin on board: insulin on board scope · case 03
Insulin on board keeps eating into the meal dose after the correction reaches zero.
Meal and correction insulin bolus with insulin on board: insulin on board scope · case 04
Insulin on board keeps eating into the meal dose after the correction reaches zero.
Meal and correction insulin bolus with insulin on board: insulin on board scope · case 05
Insulin on board keeps eating into the meal dose after the correction reaches zero.
Meal and correction insulin bolus with insulin on board: carbohydrate ratio division · case 01
A 59 g meal with a 1:10 ratio is covered with only 5 units before correction.
Meal and correction insulin bolus with insulin on board: carbohydrate ratio division · case 02
A 59 g meal with a 1:10 ratio is covered with only 5 units before correction.
Meal and correction insulin bolus with insulin on board: carbohydrate ratio division · case 03
A 59 g meal with a 1:10 ratio is covered with only 5 units before correction.
Meal and correction insulin bolus with insulin on board: carbohydrate ratio division · case 04
A 59 g meal with a 1:10 ratio is covered with only 5 units before correction.
Meal and correction insulin bolus with insulin on board: carbohydrate ratio division · case 05
A 59 g meal with a 1:10 ratio is covered with only 5 units before correction.
Meal and correction insulin bolus with insulin on board: pen half-unit rounding · case 01
A 5.9 unit calculation is delivered as 6 units.
Meal and correction insulin bolus with insulin on board: pen half-unit rounding · case 02
A 5.9 unit calculation is delivered as 6 units.
Meal and correction insulin bolus with insulin on board: pen half-unit rounding · case 03
A 5.9 unit calculation is delivered as 6 units.
Meal and correction insulin bolus with insulin on board: pen half-unit rounding · case 04
A 5.9 unit calculation is delivered as 6 units.
Meal and correction insulin bolus with insulin on board: pen half-unit rounding · case 05
A 5.9 unit calculation is delivered as 6 units.
Meal and correction insulin bolus with insulin on board: cap before pen rounding · case 01
A 7.3 unit maximum produces a 7.3 unit dose that the pen cannot deliver.
Meal and correction insulin bolus with insulin on board: cap before pen rounding · case 02
A 7.3 unit maximum produces a 7.3 unit dose that the pen cannot deliver.
Meal and correction insulin bolus with insulin on board: cap before pen rounding · case 03
A 7.3 unit maximum produces a 7.3 unit dose that the pen cannot deliver.
Meal and correction insulin bolus with insulin on board: cap before pen rounding · case 04
A 7.3 unit maximum produces a 7.3 unit dose that the pen cannot deliver.
Meal and correction insulin bolus with insulin on board: cap before pen rounding · case 05
A 7.3 unit maximum produces a 7.3 unit dose that the pen cannot deliver.
INSPECTABLE BY DESIGN
Every result has a runnable source.
Runnable implementations with recorded outputs, source hashes, and explicit contracts. Related variants share a failure mechanism and belong together in evaluation splits.
Read the methodology ↗