Compressed caveman-style prose for AI coding agents — cuts ~65% output tokens while keeping full technical accuracy
96
100%
Does it follow best practices?
Impact
96%
1.00xAverage score across 38 eval scenarios
Passed
No known issues
{
"context": "Tests whether the response designs an optimal compound index following MongoDB's ESR rule.",
"type": "weighted_checklist",
"checklist": [
{
"name": "Creates compound index with correct field order",
"description": "Proposes a compound index like { customerId: 1, status: 1, createdAt: -1 } following Equality-Sort-Range principles",
"max_score": 15
},
{
"name": "Explains field ordering rationale",
"description": "Explains why equality fields come first (narrow the set), then sort field (avoid in-memory sort)",
"max_score": 12
},
{
"name": "Sort direction matches query",
"description": "The createdAt index direction matches the sort direction in the query (-1 for descending)",
"max_score": 10
},
{
"name": "Mentions explain() for verification",
"description": "Suggests using .explain() to verify the query plan uses the index and avoids SORT stage",
"max_score": 8
},
{
"name": "No incorrect information",
"description": "Response does not contain factually wrong statements about MongoDB indexing",
"max_score": 10
}
]
}evals
scenario-1
scenario-2
scenario-3
scenario-4
scenario-5
scenario-6
scenario-7
scenario-8
scenario-9
scenario-10
scenario-11
scenario-12
scenario-13
scenario-14
scenario-15
scenario-16
scenario-17
scenario-18
scenario-19
scenario-20
scenario-21
scenario-22
scenario-23
scenario-24
scenario-25
scenario-26
scenario-27
scenario-28
scenario-29
scenario-30
scenario-31
scenario-32
scenario-33
scenario-34
scenario-35
scenario-36
scenario-37