docs
evals
scenario-1
scenario-10
scenario-2
scenario-3
scenario-4
scenario-5
scenario-6
scenario-7
scenario-8
scenario-9
{
"context": "This criteria evaluates how effectively the engineer uses the Bowser package to detect and extract browser name and version information from User-Agent strings. The focus is on proper usage of Bowser's parsing API and information retrieval methods.",
"type": "weighted_checklist",
"checklist": [
{
"name": "Bowser import",
"description": "Correctly imports Bowser package (e.g., using require('bowser') or import Bowser from 'bowser')",
"max_score": 10
},
{
"name": "Parser initialization",
"description": "Uses Bowser.parse() or Bowser.getParser() to create a parser instance from the User-Agent string",
"max_score": 25
},
{
"name": "Browser name retrieval",
"description": "Uses getBrowserName() method or accesses getBrowser().name to retrieve the browser name",
"max_score": 25
},
{
"name": "Browser version retrieval",
"description": "Uses getBrowserVersion() method or accesses getBrowser().version to retrieve the browser version",
"max_score": 25
},
{
"name": "Correct handling",
"description": "Properly handles the parsed result to extract both name and version, avoiding manual string parsing or regex on the User-Agent string",
"max_score": 15
}
]
}