---
name: undominated-judge-prompt-variance-audit
description: "Measures LLM-as-a-judge score variance, position bias, and length preference bias across automated evaluation runs. Recomputed fields must match the other inputs. A pass is not a compliance, safety, or price verdict."
license: MIT
metadata:
  author: Undominated.ai
  version: "1.1.0"
---

# Judge Prompt Variance Audit

## Description
Measures LLM-as-a-judge score variance, position bias, and length preference bias across automated evaluation runs.

## Category
`evaluation`

## Tags
`llm-judge`, `position-bias`, `eval-variance`, `evaluation`

## When to Use
Audit LLM judge evaluation harnesses to ensure scoring reproducibility and eliminate position bias.

## Operational Limitations
Evaluates observed pairwise swaps and position agreement scores.

## Verification Script
Run the deterministic verification checker:
```bash
python3 scripts/check.py examples/synthetic.json
```
Exit codes:
- `0`: Check passed and thresholds satisfied.
- `1`: Threshold breach or operational violation detected.
- `2`: Schema error or invalid usage.

## What the checker recomputes

The original threshold checks still apply:

- `swap_consistency_rate` must not be < `min_required_swap_consistency`.

Every other declared field is type-checked. A negative value is rejected except for deltas, drifts, and gains. A type check does not, by itself, prove the measurement.

`population` and `observedAt` are required. The bundled examples are synthetic fixtures, not market measurements.

```bash
python3 scripts/check.py examples/synthetic.json
python3 scripts/check.py examples/contradiction.json
python3 scripts/check.py examples/broken.json
```

Exit 0 is a pass, exit 1 is a review, and exit 2 is invalid input. A pass does not certify a provider, a regulation, a release, or a price.
