Test fixture
Bug fixes, edge conditions, and minimal patch accuracy.
The model receives the prompt (and optional system message). The run uses scorer rubric_json_metrics with the JSON configuration below. Pass/fail and partial credit are determined entirely by that scorer against the model output; no human grading.
Return JSON only with keys diagnosis, fix, tests. A header parser looks up Authorization exactly and fails when incoming headers are lowercase. Identify the bug and fix.
{
"metrics": {
"repro": {
"checks": [
{
"contains": [
"lowercase"
]
},
{
"contains": [
"Authorization"
]
},
{
"contains": [
"headers"
]
},
{
"contains": [
"fails"
]
}
]
},
"hidden": {
"checks": [
{
"contains": [
"normalize"
]
},
{
"contains": [
"case insensitive"
]
},
{
"contains": [
"authorization"
]
}
]
},
"diagnose": {
"checks": [
{
"contains": [
"case sensitive"
]
},
{
"contains": [
"header parser"
]
},
{
"contains": [
"lookup"
]
}
]
}
}
}temperature
0
max_tokens
420
timeout (s)
120
type
scored
file
debug-case-sensitive-key-v2.json