Test fixture
Code transformation while preserving behavior and intent.
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 refactor, rationale, tests. Refactor a nested workflow state machine into a transition table. Preserve invalid transition errors and side-effect hooks.
{
"metrics": {
"visible": {
"checks": [
{
"contains": [
"transition table"
]
},
{
"contains": [
"state machine"
]
},
{
"contains": [
"workflow"
]
}
]
},
"hidden": {
"checks": [
{
"contains": [
"invalid transition"
]
},
{
"contains": [
"side-effect hooks"
]
},
{
"contains": [
"errors"
]
}
]
},
"intent": {
"checks": [
{
"contains": [
"declarative"
]
},
{
"contains": [
"control flow"
]
},
{
"contains": [
"state"
]
}
]
}
}
}temperature
0
max_tokens
460
timeout (s)
120
type
scored
file
refactor-state-machine-table.json