# Run one controlled regression

> Freeze identities, establish a comparable baseline, change one surface and inspect Trial-level evidence.

---

LLMS index: [llms.txt](/harbor-self-evolving/llms.txt)

---

Snapshot the Candidate, validate the Dataset, doctor the Stack, preview Context and run diagnostic mode before promotion-eligible work when the path is new.

## You do {#you-do}

State one hypothesis and one allowed change. Keep Dataset, criteria, provider and runtime fixed unless the change deliberately creates a new baseline.

## Harbor records {#harbor-records}

Manifest digests, Context v3, model/Judge identities, Trial output, criterion Evidence, validity, coverage and governance impact.

## This does not prove {#does-not-prove}

A higher raw reward is not improvement if evidence is invalid, coverage falls or the baseline is not comparable.