Investigations with a paper trail.

An observation is useful only when you can inspect what was known then, what happened later and where the evidence falls short.

We require saved real observations, timestamped sources, independent outcome labels and an explicit review before publication. Parser corrections are listed rather than rewritten as discoveries. Shared funding alone does not identify an operator.

No independently reviewed case studies are published yet. Synthetic demonstrations and unresolved prospective observations are excluded.

Evaluation record and limits · Understand the checks