Security
Mutation-backed Solidity evaluation: vulnerability recall at a fixed false-positive budget, with patched clean negatives.
Instrumentation runs validate the scorer and pipeline deterministically; no model-driven security evidence is published yet.
| Run | Created | Mode | Evidence claim |
|---|---|---|---|
20260722T080228609Z-order-20260716 |
2026-07-22T08:02:28.609Z | instrumentation |
Validates the security fixture catalog, train/heldout leakage guard, condition ladder, deterministic scorer, enforced-decision gate, randomization, and bundle/summary reproduction. Scripted-auditor outcomes are a construction, not evidence about language models (ADR 0014). |