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 agent correctly identifies that requests.cache() does not exist instead of hallucinating an implementation. Brevity compression should not cause the agent to skip the correction and blindly implement a nonexistent API.",
"type": "weighted_checklist",
"checklist": [
{
"name": "Identifies that requests.cache() does not exist",
"description": "States that the requests library has no built-in cache() method",
"max_score": 20
},
{
"name": "Suggests a real alternative",
"description": "Recommends an actual caching approach: requests-cache, cachetools, manual caching with a dict/TTL, or functools.lru_cache",
"max_score": 15
},
{
"name": "Does not hallucinate a working requests.cache() implementation",
"description": "Does not write code that calls requests.cache() as if it were a real method",
"max_score": 20
}
]
}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