rundale-bench-judge

Community

Impartially score dialogue bundles with strict JSON.

Authordmooney
Version1.0.0
Installs0

System Documentation

What problem does it solve?

It solves the problem of consistently evaluating and ranking AI-generated dialogue by applying a shared rubric to each prompt/response pair and producing machine-usable scoring output.

Core Features & Use Cases

  • Rubric-based evaluation: Scores each candidate response across character, authenticity, language, responsiveness, and craft using a judge contract read from a system prompt file.
  • Batch bundle scoring: Reads a bundle JSON containing multiple items and returns one aggregated JSON result for the entire bundle.
  • Strict output validation: Emits JSON only, echoes the provided rubric hash unchanged, and flags issues like non-Latin script usage to support reliable downstream ingestion.

Quick Start

Ask an AI agent to run the rundale-bench-judge on a bundle JSON file from rundale-bench/.bench-queue/pending and return scores as a single JSON object only.

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: rundale-bench-judge
Download link: https://github.com/dmooney/Rundale/archive/main.zip#rundale-bench-judge

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 471,000+ vetted skills library on demand.