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 correctly distinguishes test doubles and provides appropriate examples.",
"type": "weighted_checklist",
"checklist": [
{
"name": "Correctly defines stub",
"description": "Describes a stub as returning predefined responses without behavior verification (provides canned answers)",
"max_score": 12
},
{
"name": "Correctly defines mock",
"description": "Describes a mock as verifying interactions/calls happened (behavior verification — was this method called with these args?)",
"max_score": 12
},
{
"name": "Correctly defines fake",
"description": "Describes a fake as a working lightweight implementation (in-memory database, fake SMTP server)",
"max_score": 10
},
{
"name": "Provides concrete examples for the email/DB scenario",
"description": "Maps each test double to the scenario: e.g., stub the DB for read tests, mock the email service to verify send was called, fake DB with in-memory store",
"max_score": 12
},
{
"name": "No incorrect information",
"description": "Definitions and examples are technically correct and not conflated",
"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