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 well the engineer uses pixelmatch's anti-aliasing detection capabilities to build a screenshot comparison tool. The focus is on correctly using the includeAA option to distinguish between genuine content differences and anti-aliasing artifacts.",
"type": "weighted_checklist",
"checklist": [
{
"name": "Uses pixelmatch function",
"description": "Imports and uses the pixelmatch function from the pixelmatch package to perform the image comparison",
"max_score": 10
},
{
"name": "Loads PNG images",
"description": "Uses pngjs (PNG.sync.read or similar) to load the two input PNG files and extract their image data buffers, width, and height",
"max_score": 10
},
{
"name": "Comparison with includeAA false",
"description": "Calls pixelmatch with includeAA set to false (or relies on default false) to get mismatch count excluding anti-aliased pixels",
"max_score": 25
},
{
"name": "Comparison with includeAA true",
"description": "Calls pixelmatch a second time with includeAA set to true to get mismatch count including anti-aliased pixels",
"max_score": 25
},
{
"name": "Creates output buffer",
"description": "Allocates an output buffer (Uint8Array or similar) of the correct size (width * height * 4) to receive the diff image",
"max_score": 10
},
{
"name": "Configures AA color",
"description": "Uses the aaColor option to display anti-aliased pixels in a different color (e.g., yellow) in the visual diff output",
"max_score": 10
},
{
"name": "Writes diff PNG",
"description": "Creates a PNG from the output buffer and writes it to the specified output path using pngjs",
"max_score": 10
}
]
}