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
Infusion bag run-out time across rate changes: segment boundary exhaustion · case 01
A bag that empties exactly when an infusion is paused is reported as never emptying.
Infusion bag run-out time across rate changes: segment boundary exhaustion · case 02
A bag that empties exactly when an infusion is paused is reported as never emptying.
Infusion bag run-out time across rate changes: segment boundary exhaustion · case 03
A bag that empties exactly when an infusion is paused is reported as never emptying.
Infusion bag run-out time across rate changes: segment boundary exhaustion · case 04
A bag that empties exactly when an infusion is paused is reported as never emptying.
Infusion bag run-out time across rate changes: segment boundary exhaustion · case 05
A bag that empties exactly when an infusion is paused is reported as never emptying.
Infusion bag run-out time across rate changes: volume consumed per segment · case 01
After a rate change the remaining volume is understated and run-out is predicted too early.
Infusion bag run-out time across rate changes: volume consumed per segment · case 02
After a rate change the remaining volume is understated and run-out is predicted too early.
Infusion bag run-out time across rate changes: volume consumed per segment · case 03
After a rate change the remaining volume is understated and run-out is predicted too early.
Infusion bag run-out time across rate changes: volume consumed per segment · case 04
After a rate change the remaining volume is understated and run-out is predicted too early.
Infusion bag run-out time across rate changes: volume consumed per segment · case 05
After a rate change the remaining volume is understated and run-out is predicted too early.
Infusion bag run-out time across rate changes: alarm floor at zero · case 01
A small bag shows an alarm time before the infusion started.
Infusion bag run-out time across rate changes: alarm floor at zero · case 02
A small bag shows an alarm time before the infusion started.
Infusion bag run-out time across rate changes: alarm floor at zero · case 03
A small bag shows an alarm time before the infusion started.
Infusion bag run-out time across rate changes: alarm floor at zero · case 04
A small bag shows an alarm time before the infusion started.
Infusion bag run-out time across rate changes: alarm floor at zero · case 05
A small bag shows an alarm time before the infusion started.
Infusion bag run-out time across rate changes: open-ended final segment · case 01
Bags that run out during the final rate are reported as never emptying.
Infusion bag run-out time across rate changes: open-ended final segment · case 02
Bags that run out during the final rate are reported as never emptying.
Infusion bag run-out time across rate changes: open-ended final segment · case 03
Bags that run out during the final rate are reported as never emptying.
Infusion bag run-out time across rate changes: open-ended final segment · case 04
Bags that run out during the final rate are reported as never emptying.
Infusion bag run-out time across rate changes: open-ended final segment · case 05
Bags that run out during the final rate are reported as never emptying.
Single-level extended-interval nomogram: midnight crossing · case 01
A 02:00 level after a 20:00 dose is read as 18 hours and rejected.
Single-level extended-interval nomogram: midnight crossing · case 02
A 02:00 level after a 20:00 dose is read as 18 hours and rejected.
Single-level extended-interval nomogram: midnight crossing · case 03
A 02:00 level after a 20:00 dose is read as 18 hours and rejected.
Single-level extended-interval nomogram: midnight crossing · case 04
A 02:00 level after a 20:00 dose is read as 18 hours and rejected.
Single-level extended-interval nomogram: midnight crossing · case 05
A 02:00 level after a 20:00 dose is read as 18 hours and rejected.
Single-level extended-interval nomogram: nomogram window bounds · case 01
Levels drawn exactly 6 or 14 hours after the dose are rejected.
Single-level extended-interval nomogram: nomogram window bounds · case 02
Levels drawn exactly 6 or 14 hours after the dose are rejected.
Single-level extended-interval nomogram: nomogram window bounds · case 03
Levels drawn exactly 6 or 14 hours after the dose are rejected.
Single-level extended-interval nomogram: nomogram window bounds · case 04
Levels drawn exactly 6 or 14 hours after the dose are rejected.
Single-level extended-interval nomogram: nomogram window bounds · case 05
Levels drawn exactly 6 or 14 hours after the dose are rejected.
Single-level extended-interval nomogram: time remaining on nomogram · case 01
Early levels are compared against the lowest part of each line.
Single-level extended-interval nomogram: time remaining on nomogram · case 02
Early levels are compared against the lowest part of each line.
Single-level extended-interval nomogram: time remaining on nomogram · case 03
Early levels are compared against the lowest part of each line.
Single-level extended-interval nomogram: time remaining on nomogram · case 04
Early levels are compared against the lowest part of each line.
Single-level extended-interval nomogram: time remaining on nomogram · case 05
Early levels are compared against the lowest part of each line.
Single-level extended-interval nomogram: q24h line inclusive edge · case 01
A level exactly on the q24h line is assigned q36h.
Single-level extended-interval nomogram: q24h line inclusive edge · case 02
A level exactly on the q24h line is assigned q36h.
Single-level extended-interval nomogram: q24h line inclusive edge · case 03
A level exactly on the q24h line is assigned q36h.
Single-level extended-interval nomogram: q24h line inclusive edge · case 04
A level exactly on the q24h line is assigned q36h.
Single-level extended-interval nomogram: q24h line inclusive edge · case 05
A level exactly on the q24h line is assigned q36h.
Aggregate early warning score with oxygen scale: new confusion scoring · case 01
A newly confused patient responding to voice is scored as alert.
Aggregate early warning score with oxygen scale: new confusion scoring · case 02
A newly confused patient responding to voice is scored as alert.
Aggregate early warning score with oxygen scale: new confusion scoring · case 03
A newly confused patient responding to voice is scored as alert.
Aggregate early warning score with oxygen scale: new confusion scoring · case 04
A newly confused patient responding to voice is scored as alert.
Aggregate early warning score with oxygen scale: new confusion scoring · case 05
A newly confused patient responding to voice is scored as alert.
Aggregate early warning score with oxygen scale: scale 2 air exemption · case 01
A hypercapnic-target patient saturating 95% on room air scores 2 points.
Aggregate early warning score with oxygen scale: scale 2 air exemption · case 02
A hypercapnic-target patient saturating 95% on room air scores 2 points.
Aggregate early warning score with oxygen scale: scale 2 air exemption · case 03
A hypercapnic-target patient saturating 95% on room air scores 2 points.
Aggregate early warning score with oxygen scale: scale 2 air exemption · case 04
A hypercapnic-target patient saturating 95% on room air scores 2 points.
Aggregate early warning score with oxygen scale: scale 2 air exemption · case 05
A hypercapnic-target patient saturating 95% on room air scores 2 points.
Aggregate early warning score with oxygen scale: supplemental oxygen points · case 01
Scale 2 patients on oxygen do not receive the oxygen points.
Aggregate early warning score with oxygen scale: supplemental oxygen points · case 02
Scale 2 patients on oxygen do not receive the oxygen points.
Aggregate early warning score with oxygen scale: supplemental oxygen points · case 03
Scale 2 patients on oxygen do not receive the oxygen points.
Aggregate early warning score with oxygen scale: supplemental oxygen points · case 04
Scale 2 patients on oxygen do not receive the oxygen points.
Aggregate early warning score with oxygen scale: supplemental oxygen points · case 05
Scale 2 patients on oxygen do not receive the oxygen points.
Aggregate early warning score with oxygen scale: temperature low band edge · case 01
A temperature of exactly 36.0 scores 0 instead of 1.
Aggregate early warning score with oxygen scale: temperature low band edge · case 02
A temperature of exactly 36.0 scores 0 instead of 1.
Aggregate early warning score with oxygen scale: temperature low band edge · case 03
A temperature of exactly 36.0 scores 0 instead of 1.
Aggregate early warning score with oxygen scale: temperature low band edge · case 04
A temperature of exactly 36.0 scores 0 instead of 1.
Aggregate early warning score with oxygen scale: temperature low band edge · case 05
A temperature of exactly 36.0 scores 0 instead of 1.
Aggregate early warning score with oxygen scale: single parameter trigger · case 01
A patient with aggregate 3 from two mild abnormalities is flagged as low-medium.
Aggregate early warning score with oxygen scale: single parameter trigger · case 02
A patient with aggregate 3 from two mild abnormalities is flagged as low-medium.
Aggregate early warning score with oxygen scale: single parameter trigger · case 03
A patient with aggregate 3 from two mild abnormalities is flagged as low-medium.
Aggregate early warning score with oxygen scale: single parameter trigger · case 04
A patient with aggregate 3 from two mild abnormalities is flagged as low-medium.
Aggregate early warning score with oxygen scale: single parameter trigger · case 05
A patient with aggregate 3 from two mild abnormalities is flagged as low-medium.
Sepsis screen from inflammatory criteria and bedside score: temperature criterion · case 01
A temperature of exactly 38.0 counts as an inflammatory criterion.
Sepsis screen from inflammatory criteria and bedside score: temperature criterion · case 02
A temperature of exactly 38.0 counts as an inflammatory criterion.
Sepsis screen from inflammatory criteria and bedside score: temperature criterion · case 03
A temperature of exactly 38.0 counts as an inflammatory criterion.
Sepsis screen from inflammatory criteria and bedside score: temperature criterion · case 04
A temperature of exactly 38.0 counts as an inflammatory criterion.
Sepsis screen from inflammatory criteria and bedside score: temperature criterion · case 05
A temperature of exactly 38.0 counts as an inflammatory criterion.
Sepsis screen from inflammatory criteria and bedside score: optional blood gas · case 01
Every patient without a blood gas meets the respiratory criterion.
Sepsis screen from inflammatory criteria and bedside score: optional blood gas · case 02
Every patient without a blood gas meets the respiratory criterion.
Sepsis screen from inflammatory criteria and bedside score: optional blood gas · case 03
Every patient without a blood gas meets the respiratory criterion.
Sepsis screen from inflammatory criteria and bedside score: optional blood gas · case 04
Every patient without a blood gas meets the respiratory criterion.
Sepsis screen from inflammatory criteria and bedside score: optional blood gas · case 05
Every patient without a blood gas meets the respiratory criterion.
Sepsis screen from inflammatory criteria and bedside score: white cell criterion · case 01
Band forms above 10% never count toward the criteria.
Sepsis screen from inflammatory criteria and bedside score: white cell criterion · case 02
Band forms above 10% never count toward the criteria.
Sepsis screen from inflammatory criteria and bedside score: white cell criterion · case 03
Band forms above 10% never count toward the criteria.
Sepsis screen from inflammatory criteria and bedside score: white cell criterion · case 04
Band forms above 10% never count toward the criteria.
Sepsis screen from inflammatory criteria and bedside score: white cell criterion · case 05
Band forms above 10% never count toward the criteria.
Sepsis screen from inflammatory criteria and bedside score: bedside respiratory threshold · case 01
A respiratory rate of exactly 22 does not score on the bedside tool.
Sepsis screen from inflammatory criteria and bedside score: bedside respiratory threshold · case 02
A respiratory rate of exactly 22 does not score on the bedside tool.
Sepsis screen from inflammatory criteria and bedside score: bedside respiratory threshold · case 03
A respiratory rate of exactly 22 does not score on the bedside tool.
Sepsis screen from inflammatory criteria and bedside score: bedside respiratory threshold · case 04
A respiratory rate of exactly 22 does not score on the bedside tool.
Sepsis screen from inflammatory criteria and bedside score: bedside respiratory threshold · case 05
A respiratory rate of exactly 22 does not score on the bedside tool.
Sepsis screen from inflammatory criteria and bedside score: alert precedence · case 01
Patients meeting both scores get the lower-priority SIRS alert.
Sepsis screen from inflammatory criteria and bedside score: alert precedence · case 02
Patients meeting both scores get the lower-priority SIRS alert.
Sepsis screen from inflammatory criteria and bedside score: alert precedence · case 03
Patients meeting both scores get the lower-priority SIRS alert.
Sepsis screen from inflammatory criteria and bedside score: alert precedence · case 04
Patients meeting both scores get the lower-priority SIRS alert.
Sepsis screen from inflammatory criteria and bedside score: alert precedence · case 05
Patients meeting both scores get the lower-priority SIRS alert.
Coma scale total with intubated verbal component: motor range validation · case 01
A patient who obeys commands (motor 6) is rejected as invalid.
Coma scale total with intubated verbal component: motor range validation · case 02
A patient who obeys commands (motor 6) is rejected as invalid.
Coma scale total with intubated verbal component: motor range validation · case 03
A patient who obeys commands (motor 6) is rejected as invalid.
Coma scale total with intubated verbal component: motor range validation · case 04
A patient who obeys commands (motor 6) is rejected as invalid.
Coma scale total with intubated verbal component: motor range validation · case 05
A patient who obeys commands (motor 6) is rejected as invalid.
Coma scale total with intubated verbal component: intubated verbal substitution · case 01
Intubated patients are graded one point too severe.
Coma scale total with intubated verbal component: intubated verbal substitution · case 02
Intubated patients are graded one point too severe.
Coma scale total with intubated verbal component: intubated verbal substitution · case 03
Intubated patients are graded one point too severe.
Coma scale total with intubated verbal component: intubated verbal substitution · case 04
Intubated patients are graded one point too severe.
Coma scale total with intubated verbal component: intubated verbal substitution · case 05
Intubated patients are graded one point too severe.
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 ↗