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
Unresolved source identifiers produce dangling relations · case 01
Unresolved source identifiers produce dangling relations.
Unresolved source identifiers produce dangling relations · case 02
Unresolved source identifiers produce dangling relations.
Unresolved source identifiers produce dangling relations · case 03
Unresolved source identifiers produce dangling relations.
Unresolved source identifiers produce dangling relations · case 04
Unresolved source identifiers produce dangling relations.
Unresolved source identifiers produce dangling relations · case 05
Unresolved source identifiers produce dangling relations.
Relations survive an unexposed source · case 01
Relations survive an unexposed source.
Relations survive an unexposed source · case 02
Relations survive an unexposed source.
Relations survive an unexposed source · case 03
Relations survive an unexposed source.
Relations survive an unexposed source · case 04
Relations survive an unexposed source.
Relations survive an unexposed source · case 05
Relations survive an unexposed source.
Defunct sources retain outgoing semantic links · case 01
Defunct sources retain outgoing semantic links.
Defunct sources retain outgoing semantic links · case 02
Defunct sources retain outgoing semantic links.
Defunct sources retain outgoing semantic links · case 03
Defunct sources retain outgoing semantic links.
Defunct sources retain outgoing semantic links · case 04
Defunct sources retain outgoing semantic links.
Defunct sources retain outgoing semantic links · case 05
Defunct sources retain outgoing semantic links.
Defunct targets remain resolvable · case 01
Defunct targets remain resolvable.
Defunct targets remain resolvable · case 02
Defunct targets remain resolvable.
Defunct targets remain resolvable · case 03
Defunct targets remain resolvable.
Defunct targets remain resolvable · case 04
Defunct targets remain resolvable.
Defunct targets remain resolvable · case 05
Defunct targets remain resolvable.
UTF16 range adapter counts astral text as one unit · case 01
UTF16 range adapter counts astral text as one unit.
UTF16 range adapter counts astral text as one unit · case 02
UTF16 range adapter counts astral text as one unit.
UTF16 range adapter counts astral text as one unit · case 03
UTF16 range adapter counts astral text as one unit.
UTF16 range adapter counts astral text as one unit · case 04
UTF16 range adapter counts astral text as one unit.
UTF16 range adapter counts astral text as one unit · case 05
UTF16 range adapter counts astral text as one unit.
Negative range starts are measured from zero · case 01
Negative range starts are measured from zero.
Negative range starts are measured from zero · case 02
Negative range starts are measured from zero.
Negative range starts are measured from zero · case 03
Negative range starts are measured from zero.
Negative range starts are measured from zero · case 04
Negative range starts are measured from zero.
Negative range starts are measured from zero · case 05
Negative range starts are measured from zero.
Negative range ends are measured from zero · case 01
Negative range ends are measured from zero.
Negative range ends are measured from zero · case 02
Negative range ends are measured from zero.
Negative range ends are measured from zero · case 03
Negative range ends are measured from zero.
Negative range ends are measured from zero · case 04
Negative range ends are measured from zero.
Negative range ends are measured from zero · case 05
Negative range ends are measured from zero.
Out-of-range negative starts escape normalization · case 01
Out-of-range negative starts escape normalization.
Out-of-range negative starts escape normalization · case 02
Out-of-range negative starts escape normalization.
Out-of-range negative starts escape normalization · case 03
Out-of-range negative starts escape normalization.
Out-of-range negative starts escape normalization · case 04
Out-of-range negative starts escape normalization.
Out-of-range negative starts escape normalization · case 05
Out-of-range negative starts escape normalization.
Oversized range ends are rejected instead of clamped · case 01
Oversized range ends are rejected instead of clamped.
Oversized range ends are rejected instead of clamped · case 02
Oversized range ends are rejected instead of clamped.
Oversized range ends are rejected instead of clamped · case 03
Oversized range ends are rejected instead of clamped.
Oversized range ends are rejected instead of clamped · case 04
Oversized range ends are rejected instead of clamped.
Oversized range ends are rejected instead of clamped · case 05
Oversized range ends are rejected instead of clamped.
Reversed accessible ranges are collapsed · case 01
Reversed accessible ranges are collapsed.
Reversed accessible ranges are collapsed · case 02
Reversed accessible ranges are collapsed.
Reversed accessible ranges are collapsed · case 03
Reversed accessible ranges are collapsed.
Reversed accessible ranges are collapsed · case 04
Reversed accessible ranges are collapsed.
Reversed accessible ranges are collapsed · case 05
Reversed accessible ranges are collapsed.
Range start rounds into the next text character · case 01
Range start rounds into the next text character.
Range start rounds into the next text character · case 02
Range start rounds into the next text character.
Range start rounds into the next text character · case 03
Range start rounds into the next text character.
Range start rounds into the next text character · case 04
Range start rounds into the next text character.
Range start rounds into the next text character · case 05
Range start rounds into the next text character.
Range end drops a partially addressed astral character · case 01
Range end drops a partially addressed astral character.
Range end drops a partially addressed astral character · case 02
Range end drops a partially addressed astral character.
Range end drops a partially addressed astral character · case 03
Range end drops a partially addressed astral character.
Range end drops a partially addressed astral character · case 04
Range end drops a partially addressed astral character.
Range end drops a partially addressed astral character · case 05
Range end drops a partially addressed astral character.
Platform text ranges expose carriage-return duplicates · case 01
Platform text ranges expose carriage-return duplicates.
Platform text ranges expose carriage-return duplicates · case 02
Platform text ranges expose carriage-return duplicates.
Platform text ranges expose carriage-return duplicates · case 03
Platform text ranges expose carriage-return duplicates.
Platform text ranges expose carriage-return duplicates · case 04
Platform text ranges expose carriage-return duplicates.
Platform text ranges expose carriage-return duplicates · case 05
Platform text ranges expose carriage-return duplicates.
Embedded NUL terminates accessible text content · case 01
Embedded NUL terminates accessible text content.
Embedded NUL terminates accessible text content · case 02
Embedded NUL terminates accessible text content.
Embedded NUL terminates accessible text content · case 03
Embedded NUL terminates accessible text content.
Embedded NUL terminates accessible text content · case 04
Embedded NUL terminates accessible text content.
Embedded NUL terminates accessible text content · case 05
Embedded NUL terminates accessible text content.
CRLF consumes two places in a protected value cap · case 01
CRLF consumes two places in a protected value cap.
CRLF consumes two places in a protected value cap · case 02
CRLF consumes two places in a protected value cap.
CRLF consumes two places in a protected value cap · case 03
CRLF consumes two places in a protected value cap.
CRLF consumes two places in a protected value cap · case 04
CRLF consumes two places in a protected value cap.
CRLF consumes two places in a protected value cap · case 05
CRLF consumes two places in a protected value cap.
Single-line accessible value retains line breaks · case 01
Single-line accessible value retains line breaks.
Single-line accessible value retains line breaks · case 02
Single-line accessible value retains line breaks.
Single-line accessible value retains line breaks · case 03
Single-line accessible value retains line breaks.
Single-line accessible value retains line breaks · case 04
Single-line accessible value retains line breaks.
Single-line accessible value retains line breaks · case 05
Single-line accessible value retains line breaks.
A zero length cap leaks the entire protected value · case 01
A zero length cap leaks the entire protected value.
A zero length cap leaks the entire protected value · case 02
A zero length cap leaks the entire protected value.
A zero length cap leaks the entire protected value · case 03
A zero length cap leaks the entire protected value.
A zero length cap leaks the entire protected value · case 04
A zero length cap leaks the entire protected value.
A zero length cap leaks the entire protected value · case 05
A zero length cap leaks the entire protected value.
The value cap leaks one extra character · case 01
The value cap leaks one extra character.
The value cap leaks one extra character · case 02
The value cap leaks one extra character.
The value cap leaks one extra character · case 03
The value cap leaks one extra character.
The value cap leaks one extra character · case 04
The value cap leaks one extra character.
The value cap leaks one extra character · case 05
The value cap leaks one extra character.
Either reveal or permission independently unmasks a value · case 01
Either reveal or permission independently unmasks a value.
Either reveal or permission independently unmasks a value · case 02
Either reveal or permission independently unmasks a value.
Either reveal or permission independently unmasks a value · case 03
Either reveal or permission independently unmasks a value.
Either reveal or permission independently unmasks a value · case 04
Either reveal or permission independently unmasks a value.
Either reveal or permission independently unmasks a value · case 05
Either reveal or permission independently unmasks a value.
An unprotected control is unnecessarily masked · case 01
An unprotected control is unnecessarily masked.
An unprotected control is unnecessarily masked · case 02
An unprotected control is unnecessarily masked.
An unprotected control is unnecessarily masked · case 03
An unprotected control is unnecessarily masked.
An unprotected control is unnecessarily masked · case 04
An unprotected control is unnecessarily masked.
An unprotected control is unnecessarily masked · case 05
An unprotected control is unnecessarily masked.
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 ↗