# Neon Durable Guitar Candidate Report

- date: 2026-05-25
- target: https://main.dlwavl00q582x.amplifyapp.com
- lilarcade_pr: 522
- lilarcade_deploy_commit: 060767a0157b1a9da1de91bb96c73ad324166079
- lilarcade_amplify_job: 700
- status: ready_for_promotion_review

## Claim

A transient Neon `guitar -0.05` claim candidate can be promoted into a durable deployed override only after objective receipts pass, an approval-surrogate packet is captured, a durable patch plan is generated, and the deployed current target proves the override is active without deterministic audio/app regressions.

## Promotion Proof

The production promotion batch ran against the deployed Neon target with the requested intent:

```text
turn the guitar part down a little
```

The batch passed:

- fast mix health
- mobile trainer layout
- playback sync
- deep exploration over 2 songs, 4 parts, and 4 windows
- narrowed guitar/down claim-candidate review
- approval-surrogate candidate application
- durable patch plan generation
- pre-patch current-target audit for already-active overrides

The review packet recommended `guitar -0.05`. The approved packet captured `candidate_applied_for_listening_review` with `approval_mode: mixing_canon_surrogate`.

## Durable Patch

The durable patch plan applied this override:

```json
{
  "id": "monkberry-moon-delight-tab-guitar-065-to-06-approved-candidate",
  "status": "active",
  "mixerLevels": {
    "guitar": 0.6
  }
}
```

The patch keeps the approval boundary explicit: this is a proof-backed candidate handoff, not proof of subjective mix quality.

## Post-Deploy Current-Target Proof

After LilArcade PR #522 deployed, the live current-target proof passed with two active durable overrides and zero findings:

- chord preservation override: `chord: 0.16`
- guitar candidate override: `guitar: 0.6`

The guitar override proof reported:

- rms: `0.1058`
- peak: `0.7666`
- clipping: `false`
- headroomDb: `2.31`
- lowLevel: `false`

## Boundary

This proves the candidate was safely promoted through deterministic app/audio receipts and is active on the deployed target. It does not prove the mix sounds better; listening review remains the taste gate.
