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 tenant isolation where queries filter by resource id but forget tenant_id in some paths.
{
"metrics": {
"correct": {
"checks": [
{
"contains": [
"tenant"
]
},
{
"contains": [
"tenant_id"
]
},
{
"contains": [
"isolation"
]
},
{
"contains": [
"query"
]
},
{
"contains": [
"access control"
]
}
]
},
"hidden": {
"checks": [
{
"contains": [
"row level"
]
},
{
"contains": [
"ownership"
]
},
{
"contains": [
"deny"
]
},
{
"contains": [
"test"
]
},
{
"contains": [
"cross-tenant"
]
}
]
}
}
}temperature
0
max_tokens
420
timeout (s)
120
type
scored
file
sec-tenant-isolation-checker.json