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. Session cookies are set with SameSite=None for every environment even though most flows do not require cross-site cookies.
{
"metrics": {
"correct": {
"checks": [
{
"contains": [
"SameSite=None"
]
},
{
"contains": [
"session cookies"
]
},
{
"contains": [
"cross-site"
]
}
]
},
"hidden": {
"checks": [
{
"contains": [
"SameSite=Lax"
]
},
{
"contains": [
"Secure"
]
},
{
"contains": [
"CSRF"
]
}
]
}
}
}temperature
0
max_tokens
460
timeout (s)
120
type
scored
file
sec-samesite-cookie-policy.json