AAegis

Trust

Claims & Methodology

Available now

Every public claim maps to an implementation test or a named synthetic evaluation.

Leak test

The project’s own 12-prompt test planted 35 entities and observed 0 reaching Upstream. It is not an adversarial audit.

Detection evaluation

On a 220-sentence synthetic project-authored set, PERSON recall measured 96.9% and precision 61.9%.

Performance

Measure rule and model overhead on your own deployment using bench_scrub.py; results depend on hardware and workload.