CtrlK
BlogDocsLog inGet started
Tessl Logo

juliusbrussee/caveman

Compressed caveman-style prose for AI coding agents — cuts ~65% output tokens while keeping full technical accuracy

96

1.00x
Quality

100%

Does it follow best practices?

Impact

96%

1.00x

Average score across 38 eval scenarios

SecuritybySnyk

Passed

No known issues

Overview
Quality
Evals
Security
Files

criteria.jsonevals/scenario-14/

{
  "context": "Tests whether the response correctly explains the GraphQL N+1 problem and provides the standard solution.",
  "type": "weighted_checklist",
  "checklist": [
    {
      "name": "Explains why N+1 happens in GraphQL",
      "description": "Explains that each field resolver runs independently, so the author resolver fires once per post",
      "max_score": 12
    },
    {
      "name": "Recommends DataLoader or batching",
      "description": "Suggests DataLoader (or equivalent batching/coalescing library) to batch author lookups into a single query",
      "max_score": 15
    },
    {
      "name": "Explains DataLoader mechanism",
      "description": "Describes how DataLoader collects IDs during a tick and makes one batched query (WHERE id IN (...))",
      "max_score": 10
    },
    {
      "name": "Mentions per-request DataLoader instances",
      "description": "Notes that DataLoader instances should be created per-request to avoid stale cache across requests",
      "max_score": 8
    },
    {
      "name": "No incorrect information",
      "description": "Response does not contain factually wrong statements about GraphQL resolution or DataLoader",
      "max_score": 10
    }
  ]
}

evals

README.md

SKILL.md

tile.json