Skills
.Work. You
Rest

cheat-score-blind

Community

Blind rubric scoring for isolated evaluation.

Software Engineering#calibration#rubric#json-output#script-analysis#blind-scoring#internal-tool
AuthorJason5330
Version1.0.0
Installs0

System Documentation

What problem does it solve?

This internal sub-agent provides a fully isolated, multi-dimensional rubric score for a given script by consuming only script_path and rubric_notes_path, ensuring no leakage of actual outcomes or state data into the main conversation.

Core Features & Use Cases

  • Reads only the specified inputs (script_path and rubric_notes_path) to produce an objective JSON score across N dimensions.
  • Enforces strict output format with per-dimension scores (0-5) and a confidence level, suitable for calibration workflows.
  • Used by cheat-score, cheat-predict, and cheat-bump to generate a clean anchor for model evaluation without data contamination.

Quick Start

Spawn cheat-score-blind with a given script_path and rubric_notes_path to obtain a strict JSON score.

Dependency Matrix

Required Modules

None required

Components

Standard package

💻 Claude Code Installation

Recommended: Let Claude install automatically. Simply copy and paste the text below to Claude Code.

Please help me install this Skill:
Name: cheat-score-blind
Download link: https://github.com/Jason5330/ai-self-eval/archive/main.zip#cheat-score-blind

Please download this .zip file, extract it, and install it in the .claude/skills/ directory.
View Source Repository

Agent Skills Search Helper

Install a tiny helper to your Agent, search and equip skill from 620,000+ vetted skills library on demand.