Files
gstack/evals/parity/regressions/pr-2023.json
T

24 lines
701 B
JSON

{
"pr": 2023,
"url": "https://github.com/garrytan/gstack/pull/2023",
"title": "Label single-model autoplan output honestly",
"targets": [
"autoplan"
],
"anchor": "GSTACK2_FIX_2023_SINGLE_VOICE_LABELS",
"body": "### Single-voice labeling\n\nWhen only one model produced a review row, label it **Claude-only** or **Codex-only** and print a visible single-voice banner. Never describe a one-model result as consensus, agreement, or cross-model validation.",
"regression": {
"input": {
"available_models": [
"Codex"
],
"review_rows": 4
},
"expected": {
"label": "Codex-only",
"banner": true,
"consensus_claim": false
}
}
}