Test fixture
Secure code changes, vulnerability recognition, and safe defaults.
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 verdict, risk, mitigation. Review refresh token handling that never rotates tokens and does not detect replay.
{
"metrics": {
"correct": {
"checks": [
{
"contains": [
"refresh token"
]
},
{
"contains": [
"rotation"
]
},
{
"contains": [
"replay"
]
},
{
"contains": [
"revoke"
]
},
{
"contains": [
"session"
]
}
]
},
"hidden": {
"checks": [
{
"contains": [
"reuse detection"
]
},
{
"contains": [
"family"
]
},
{
"contains": [
"hash token"
]
},
{
"contains": [
"expire"
]
},
{
"contains": [
"invalidate"
]
}
]
}
}
}temperature
0
max_tokens
420
timeout (s)
120
type
scored
file
sec-refresh-token-rotation.json