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. An OAuth public client stores the PKCE code_verifier in localStorage and reuses the same verifier across login attempts.
{
"metrics": {
"correct": {
"checks": [
{
"contains": [
"PKCE"
]
},
{
"contains": [
"code_verifier"
]
},
{
"contains": [
"reuse"
]
}
]
},
"hidden": {
"checks": [
{
"contains": [
"per request"
]
},
{
"contains": [
"state binding"
]
},
{
"contains": [
"localStorage risk"
]
}
]
}
}
}temperature
0
max_tokens
560
timeout (s)
120
type
scored
file
sec-oauth-pkce-verifier.json