ResidueGuard
Field-level data integrity that detects, locates, and auto-corrects single-byte errors in tabular data — three bytes per field, fully air-gappable.
What it is.
Most data-integrity tooling tells you a row was corrupted; ResidueGuard tells you which field, and when the corruption is within the recoverable envelope, fixes it in place.
Three bytes of side-channel state per protected field. No checksum-of-the-checksum. No hashing the whole row. The detection signal is local to the field, which is what makes pinpoint correction possible.
Engineered for environments where shipping data to a cloud verifier is not an option — finance reconciliation, regulated records, classified pipelines. The library runs entirely in-process and produces no telemetry.
What it does.
- 100% detection of single-field modifications across 50,000+ corruption-scenario trials.
- ≈89% auto-correct rate on single-byte errors, with the field identified before correction.
- 3 bytes per field overhead — fixed-cost, not row-proportional.
- Air-gap deployable — no cloud, no callback, no telemetry.
- SDK in Python and C for ETL, batch, and embedded integration paths.
Who it is for.
Finance reconciliation teams, regulated-data pipelines, classified-record systems, and industrial historians — wherever a corrupted field cannot be retried out of, but also cannot be silently propagated downstream.
Contact for details.
Datasheet, integration scope, and pricing under NDA. Reference this product in your subject line.
contact@newleaftoolsllc.com