Context
Across structural, extraction, and bridge trials, apparent success sometimes hid missing evidence or an untested integration boundary.
VIS-FAIL-01METHOD VERIFIEDMethod note
Preserving Failure as a Product Requirement
A passing correction should not erase the failure that made the correction necessary.
Governing rule
Preserve the original failure, narrow the correction, and rerun the boundary.
Current state
Original failures and corrected reruns remain paired in the record.
Current evidence
What the record supports.
- Initial structural result
- 7 / 35
- Corrected structural result
- 35 / 35
- Extraction cases
- 6 / 6
- Original failures remain visible
- Corrections are scoped to the failed boundary
- Promotion stays blocked when evidence is missing
Public evidence files
Inspect the record.
Public-safe derivatives only. Source availability does not broaden the claim stated on this page.
Claim boundary
What remains unearned.
- Not exhaustive validation
- Not a production-security claim
- Direct component success does not prove integration
Next evidence
Turn the paired records into a public regression package.