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. A redirect URL is taken directly from a query parameter without any allowlist validation before issuing the redirect.
{
"metrics": {
"correct": {
"checks": [
{
"contains": [
"open redirect"
]
},
{
"contains": [
"allowlist"
]
},
{
"contains": [
"phishing"
]
},
{
"contains": [
"validate URL"
]
},
{
"contains": [
"whitelist"
]
}
]
},
"hidden": {
"checks": [
{
"contains": [
"relative path"
]
},
{
"contains": [
"URL parsing"
]
},
{
"contains": [
"hostname"
]
},
{
"contains": [
"token-based"
]
},
{
"contains": [
"user confirmation"
]
}
]
}
}
}temperature
0
max_tokens
420
timeout (s)
120
type
scored
file
sec-open-redirect.json