score-eval

Score text-based evaluation diffs against a predefined rubric.

83|14|Updated Jan 20, 2026
One-click install
npx skills add https://github.com/neondatabase/agent-skills --skill score-eval
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: score-eval
Source: https://github.com/neondatabase/agent-skills/tree/main/evals/neon-postgres-egress-optimizer/.claude/skills/score-eval
Command: npx skills add https://github.com/neondatabase/agent-skills --skill score-eval

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

This Skill automates the process of scoring evaluation diffs against a predefined rubric, simplifying the analysis and scoring of complex evaluation tasks.

Core Features & Use Cases

  • Diff Analysis: Reads and analyzes the diff file provided.
  • Rubric Matching: Compares the diff against an eval rubric.
  • Result Generation: Appends findings to a results.csv file.
  • Use Case: Ideal for automating the scoring of code reviews, project evaluations, or any scenario where a rubric-based scoring system is applied.

Quick Start

Run the score-eval skill with the path to your diff file and the rubric file to get a scored evaluation.

Frequently Asked Questions about score-eval

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I automate scoring code review diffs against a rubric?

Automating diff scoring against a rubric requires reading text-based code diffs, comparing them to structured evaluation criteria, and appending the evaluated results to a results.csv file.

What is the best way to evaluate project diffs using a predefined rubric?

Evaluating project diffs using a predefined rubric involves analyzing the text diff file, matching the changes against your structured scoring criteria, and generating an automated scored evaluation output.

How do I use a rubric to score an evaluation diff file?

To score an evaluation diff file, provide the script with the path to your diff file and the rubric file; the script performs diff analysis and appends the findings to a results.csv file.

Can I use this rubric matching approach for automated code reviews?

Yes, rubric matching for automated code reviews operates on text-based diffs and structured criteria, making it suitable for project evaluations and similar automated scoring workflows.

How does diff analysis and rubric matching generate evaluation results?

Diff analysis reads the provided text diff, rubric matching compares it against structured criteria, and result generation appends the scored findings to a results.csv file for tracking.

What format do the evaluation results get saved to when scoring diffs?

When scoring diffs against a predefined rubric, the evaluation results are generated and appended to a results.csv file, providing a structured text-based output for your scoring workflow.