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 produces a correct multi-stage Dockerfile with proper separation of build and runtime stages.",
"type": "weighted_checklist",
"checklist": [
{
"name": "Has build stage",
"description": "Defines a build stage (FROM ... AS build or similar) that installs dependencies and compiles TypeScript",
"max_score": 12
},
{
"name": "Has separate runtime stage",
"description": "Defines a second stage that copies only the compiled output, not the full build environment",
"max_score": 12
},
{
"name": "Uses slim or alpine base for runtime",
"description": "The runtime stage uses a minimal base image (node:*-slim, node:*-alpine, or distroless)",
"max_score": 10
},
{
"name": "Copies only production artifacts",
"description": "The runtime stage copies compiled JS and production node_modules (or just compiled output), not TypeScript source or devDependencies",
"max_score": 12
},
{
"name": "Runs TypeScript compilation",
"description": "Build stage includes a step to compile TypeScript (npx tsc, npm run build, or similar)",
"max_score": 10
},
{
"name": "Valid Dockerfile syntax",
"description": "The Dockerfile is syntactically valid and would build without errors given standard project files",
"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